AI & Data Science — Expert Programme
Expert-level practical training in AI & Data Science, structured into 24 sequenced modules with hands-on exercises, guided projects and a capstone. Curriculum and course details are drafts pending academic review.
Book a Free Demo →Course overview
Expert-level practical training in AI & Data Science, structured into 24 sequenced modules with hands-on exercises, guided projects and a capstone. Curriculum and course details are drafts pending academic review.
Who can join?
Learners seeking structured, practical skills in AI & Data Science; final eligibility and guidance will be confirmed by the training team.
Prerequisites
Basic computer literacy. Module-specific prerequisites are covered in the course roadmap; final entry criteria require trainer review.
Practical projects
Hands-on practical exercises across 24 modules in AI & Data Science, followed by an integrated portfolio/capstone project. Final projects, assessment rubrics and industry relevance require trainer approval.
Classroom curriculum
01Computer & Developer Foundations7.5 hrs
Learn computer & developer foundations through structured practical training at IDM Technologies, Vallioor. Cover computing and professional workflow, files, paths and command line, editor and browser developer tools, and documentation and debugging habits. Build a working software feature or technical demonstration, practise evidence-based review, and demonstrate job-relevant skills. This is a draft module outline, subject to trainer approval.
Topics and subtopics
Computing and professional workflow: Operating systems; files and extensions; project naming; safe sample data Files, paths and command line: Absolute/relative paths; cd, dir/ls, mkdir; safe operations Editor and browser developer tools: VS Code workspace; extensions; search; browser inspector; console Documentation and debugging habits: Official docs; version checks; minimal reproduction; error logs Workstation assessment: Folder organization; reproducibility; documentation; submission
Learning outcomes
Explain computing and professional workflow and distinguish correct use from a common misconception. Explain files, paths and command line and distinguish correct use from a common misconception. Explain editor and browser developer tools and distinguish correct use from a common misconception. Explain documentation and debugging habits and distinguish correct use from a common misconception. Explain workstation assessment and distinguish correct use from a common misconception. Apply Computing and professional workflow, Files, paths and command line, Editor and browser developer tools to create a working software feature or technical demonstration with independently checkable evidence. Evaluate the computer & developer foundations deliverable against stated requirements, safety or quality criteria, and document corrections.
Practical activities
Workshop 1 — Computing and professional workflow: Create the developer workspace and three project folders. Submit: source code, tests, setup notes and a short walkthrough. Workshop 2 — Files, paths and command line: Navigate and create nested project directories from the terminal. Submit: source code, tests, setup notes and a short walkthrough. Workshop 3 — Editor and browser developer tools: Configure editor and inspect a sample web page. Submit: source code, tests, setup notes and a short walkthrough. Workshop 4 — Documentation and debugging habits: Reproduce a setup task and record troubleshooting steps. Submit: source code, tests, setup notes and a short walkthrough. Workshop 5 — Workstation assessment: Demonstrate a clean reproducible workspace. Submit: source code, tests, setup notes and a short walkthrough. Independent task 1 — Computing and professional workflow: Draw the initial requirements, user personas and one-page scope for all three portfolio projects. Independent task 2 — Files, paths and command line: Recreate the project directories entirely with commands and capture the command log. Independent task 3 — Editor and browser developer tools: Prepare README files and a development-environment checklist for each portfolio project. Independent task 4 — Documentation and debugging habits: Write user stories and acceptance criteria for one feature in each project. Independent task 5 — Workstation assessment: Submit all three project charters, folder trees and setup notes.
02Python Programming Fundamentals7.5 hrs
Learn python programming fundamentals through structured practical training at IDM Technologies, Vallioor. Cover python setup and syntax, operators and branching, loops and patterns, and functions and scope. Build a working software feature or technical demonstration, practise evidence-based review, and demonstrate job-relevant skills. This is a draft module outline, subject to trainer approval.
Topics and subtopics
Python setup and syntax: Interpreter; pip; virtual environment basics; variables; print/input Operators and branching: Comparison; logical operators; if/elif/else Loops and patterns: for/while; range; break/continue; nested loops Functions and scope: Parameters; return; defaults; local scope Python fundamentals assessment: Trace/debug; input safety; edge cases
Learning outcomes
Explain python setup and syntax and distinguish correct use from a common misconception. Explain operators and branching and distinguish correct use from a common misconception. Explain loops and patterns and distinguish correct use from a common misconception. Explain functions and scope and distinguish correct use from a common misconception. Explain python fundamentals assessment and distinguish correct use from a common misconception. Apply Python setup and syntax, Operators and branching, Loops and patterns to create a working software feature or technical demonstration with independently checkable evidence. Evaluate the python programming fundamentals deliverable against stated requirements, safety or quality criteria, and document corrections.
Practical activities
Workshop 1 — Python setup and syntax: Run Python calculations and types demo. Submit: source code, tests, setup notes and a short walkthrough. Workshop 2 — Operators and branching: Implement validation and decision problems. Submit: source code, tests, setup notes and a short walkthrough. Workshop 3 — Loops and patterns: Solve loop exercises with trace tables. Submit: source code, tests, setup notes and a short walkthrough. Workshop 4 — Functions and scope: Refactor procedural scripts to functions. Submit: source code, tests, setup notes and a short walkthrough. Workshop 5 — Python fundamentals assessment: Complete unseen coding problems independently. Submit: source code, tests, setup notes and a short walkthrough. Independent task 1 — Python setup and syntax: Write a simple command-line admission fee calculator. Independent task 2 — Operators and branching: Write job eligibility checks based on sample criteria. Independent task 3 — Loops and patterns: Process a list of booking requests and summarize counts. Independent task 4 — Functions and scope: Create reusable helpers for fee, job and service calculations. Independent task 5 — Python fundamentals assessment: Submit a tested command-line utility with examples and documentation.
03Python OOP & Packages20 hrs
Learn python oop & packages through structured practical training at IDM Technologies, Vallioor. Cover classes, inheritance, environments, and packages. Build a working software feature or technical demonstration, practise evidence-based review, and demonstrate job-relevant skills. This is a draft module outline, subject to trainer approval.
Topics and subtopics
Classes: Classes: key definitions and purpose; Classes: workflow and applied decisions in software development; Classes: demonstration, error review and independent verification Inheritance: inheritance: key definitions and purpose; inheritance: workflow and applied decisions in software development; inheritance: demonstration, error review and independent verification Environments: environments: key definitions and purpose; environments: workflow and applied decisions in software development; environments: demonstration, error review and independent verification Packages: packages: key definitions and purpose; packages: workflow and applied decisions in software development; packages: demonstration, error review and independent verification Testing: testing: test cases, expected results and fixtures; testing: negative cases, regression and bug reports; testing: write and run repeatable validation checks
Learning outcomes
Explain classes and distinguish correct use from a common misconception. Explain inheritance and distinguish correct use from a common misconception. Explain environments and distinguish correct use from a common misconception. Explain packages and distinguish correct use from a common misconception. Explain testing and distinguish correct use from a common misconception. Apply Classes, Inheritance, Environments to create a working software feature or technical demonstration with independently checkable evidence. Evaluate the python oop & packages deliverable against stated requirements, safety or quality criteria, and document corrections.
Practical activities
Workshop 1 — Classes: produce a focused example applying classes to a realistic brief; compare the initial result with a corrected result and record the reasoning. Submit: source code, tests, setup notes and a short walkthrough. Workshop 2 — Inheritance: produce a focused example applying inheritance to a realistic brief; compare the initial result with a corrected result and record the reasoning. Submit: source code, tests, setup notes and a short walkthrough. Workshop 3 — Environments: produce a focused example applying environments to a realistic brief; compare the initial result with a corrected result and record the reasoning. Submit: source code, tests, setup notes and a short walkthrough. Workshop 4 — Packages: produce a focused example applying packages to a realistic brief; compare the initial result with a corrected result and record the reasoning. Submit: source code, tests, setup notes and a short walkthrough. Workshop 5 — Testing: produce a focused example applying testing to a realistic brief; compare the initial result with a corrected result and record the reasoning. Submit: source code, tests, setup notes and a short walkthrough. Applied mini-project — Python OOP & Packages: assemble a working software feature or technical demonstration combining Classes, Inheritance, Environments; document inputs, outputs and constraints. Review and peer demonstration — Python OOP & Packages: identify two likely mistakes, run a validation checklist, and explain improvements with evidence.
04Data Preparation & Quality16 hrs
Learn data preparation & quality through structured practical training at IDM Technologies, Vallioor. Cover data types, profiling, cleansing, and missing data. Build a reproducible analysis workbook or query notebook, practise evidence-based review, and demonstrate job-relevant skills. This is a draft module outline, subject to trainer approval.
Topics and subtopics
Data types: Data types: key definitions and purpose; Data types: workflow and applied decisions in data analytics; Data types: demonstration, error review and independent verification Profiling: profiling: key definitions and purpose; profiling: workflow and applied decisions in data analytics; profiling: demonstration, error review and independent verification Cleansing: cleansing: key definitions and purpose; cleansing: workflow and applied decisions in data analytics; cleansing: demonstration, error review and independent verification Missing data: missing data: key definitions and purpose; missing data: workflow and applied decisions in data analytics; missing data: demonstration, error review and independent verification Validation: validation: key definitions and purpose; validation: workflow and applied decisions in data analytics; validation: demonstration, error review and independent verification
Learning outcomes
Explain data types and distinguish correct use from a common misconception. Explain profiling and distinguish correct use from a common misconception. Explain cleansing and distinguish correct use from a common misconception. Explain missing data and distinguish correct use from a common misconception. Explain validation and distinguish correct use from a common misconception. Apply Data types, Profiling, Cleansing to create a reproducible analysis workbook or query notebook with independently checkable evidence. Evaluate the data preparation & quality deliverable against stated requirements, safety or quality criteria, and document corrections.
Practical activities
Workshop 1 — Data types: produce a focused example applying data types to a realistic brief; compare the initial result with a corrected result and record the reasoning. Submit: input dataset, transformation checks, chart or query results and interpretation. Workshop 2 — Profiling: produce a focused example applying profiling to a realistic brief; compare the initial result with a corrected result and record the reasoning. Submit: input dataset, transformation checks, chart or query results and interpretation. Workshop 3 — Cleansing: produce a focused example applying cleansing to a realistic brief; compare the initial result with a corrected result and record the reasoning. Submit: input dataset, transformation checks, chart or query results and interpretation. Workshop 4 — Missing data: produce a focused example applying missing data to a realistic brief; compare the initial result with a corrected result and record the reasoning. Submit: input dataset, transformation checks, chart or query results and interpretation. Workshop 5 — Validation: produce a focused example applying validation to a realistic brief; compare the initial result with a corrected result and record the reasoning. Submit: input dataset, transformation checks, chart or query results and interpretation. Applied mini-project — Data Preparation & Quality: assemble a reproducible analysis workbook or query notebook combining Data types, Profiling, Cleansing; document inputs, outputs and constraints. Review and peer demonstration — Data Preparation & Quality: identify two likely mistakes, run a validation checklist, and explain improvements with evidence.
05Statistics for Data Work20 hrs
Learn statistics for data work through structured practical training at IDM Technologies, Vallioor. Cover distributions, variation, sampling, and probability. Build a reproducible analysis workbook or query notebook, practise evidence-based review, and demonstrate job-relevant skills. This is a draft module outline, subject to trainer approval.
Topics and subtopics
Distributions: Distributions: key definitions and purpose; Distributions: workflow and applied decisions in data analytics; Distributions: demonstration, error review and independent verification Variation: variation: key definitions and purpose; variation: workflow and applied decisions in data analytics; variation: demonstration, error review and independent verification Sampling: sampling: key definitions and purpose; sampling: workflow and applied decisions in data analytics; sampling: demonstration, error review and independent verification Probability: probability: key definitions and purpose; probability: workflow and applied decisions in data analytics; probability: demonstration, error review and independent verification Hypothesis basics: hypothesis basics: key definitions and purpose; hypothesis basics: workflow and applied decisions in data analytics; hypothesis basics: demonstration, error review and independent verification
Learning outcomes
Explain distributions and distinguish correct use from a common misconception. Explain variation and distinguish correct use from a common misconception. Explain sampling and distinguish correct use from a common misconception. Explain probability and distinguish correct use from a common misconception. Explain hypothesis basics and distinguish correct use from a common misconception. Apply Distributions, Variation, Sampling to create a reproducible analysis workbook or query notebook with independently checkable evidence. Evaluate the statistics for data work deliverable against stated requirements, safety or quality criteria, and document corrections.
Practical activities
Workshop 1 — Distributions: produce a focused example applying distributions to a realistic brief; compare the initial result with a corrected result and record the reasoning. Submit: input dataset, transformation checks, chart or query results and interpretation. Workshop 2 — Variation: produce a focused example applying variation to a realistic brief; compare the initial result with a corrected result and record the reasoning. Submit: input dataset, transformation checks, chart or query results and interpretation. Workshop 3 — Sampling: produce a focused example applying sampling to a realistic brief; compare the initial result with a corrected result and record the reasoning. Submit: input dataset, transformation checks, chart or query results and interpretation. Workshop 4 — Probability: produce a focused example applying probability to a realistic brief; compare the initial result with a corrected result and record the reasoning. Submit: input dataset, transformation checks, chart or query results and interpretation. Workshop 5 — Hypothesis basics: produce a focused example applying hypothesis basics to a realistic brief; compare the initial result with a corrected result and record the reasoning. Submit: input dataset, transformation checks, chart or query results and interpretation. Applied mini-project — Statistics for Data Work: assemble a reproducible analysis workbook or query notebook combining Distributions, Variation, Sampling; document inputs, outputs and constraints. Review and peer demonstration — Statistics for Data Work: identify two likely mistakes, run a validation checklist, and explain improvements with evidence.
06Linear Algebra for Machine Learning20 hrs
Learn linear algebra for machine learning through structured practical training at IDM Technologies, Vallioor. Cover vectors, matrices, transforms, and eigen ideas. Build a documented model experiment or AI workflow, practise evidence-based review, and demonstrate job-relevant skills. This is a draft module outline, subject to trainer approval.
Topics and subtopics
Vectors: Vectors: scalar versus vector quantities; coordinates and dimensions; Vectors: vector addition, magnitude and dot product; Vectors: feature-vector example and shape checks Matrices: matrices: rows, columns, dimensions and indexing; matrices: matrix multiplication, transpose and transformations; matrices: small matrix computation with shape verification Transforms: transforms: key definitions and purpose; transforms: workflow and applied decisions in ai data science; transforms: demonstration, error review and independent verification Eigen ideas: eigen ideas: eigenvalues, eigenvectors and intuition; eigen ideas: variance directions and dimensionality reduction; eigen ideas: interpret a simple two-dimensional example
Learning outcomes
Explain vectors and distinguish correct use from a common misconception. Explain matrices and distinguish correct use from a common misconception. Explain transforms and distinguish correct use from a common misconception. Explain eigen ideas and distinguish correct use from a common misconception. Apply Vectors, Matrices, Transforms to create a documented model experiment or AI workflow with independently checkable evidence. Evaluate the linear algebra for machine learning deliverable against stated requirements, safety or quality criteria, and document corrections.
Practical activities
Workshop 1 — Vectors: produce a focused example applying vectors to a realistic brief; compare the initial result with a corrected result and record the reasoning. Submit: data provenance, evaluation criteria, limitations and reproducibility. Workshop 2 — Matrices: produce a focused example applying matrices to a realistic brief; compare the initial result with a corrected result and record the reasoning. Submit: data provenance, evaluation criteria, limitations and reproducibility. Workshop 3 — Transforms: produce a focused example applying transforms to a realistic brief; compare the initial result with a corrected result and record the reasoning. Submit: data provenance, evaluation criteria, limitations and reproducibility. Workshop 4 — Eigen ideas: produce a focused example applying eigen ideas to a realistic brief; compare the initial result with a corrected result and record the reasoning. Submit: data provenance, evaluation criteria, limitations and reproducibility. Applied mini-project — Linear Algebra for Machine Learning: assemble a documented model experiment or AI workflow combining Vectors, Matrices, Transforms; document inputs, outputs and constraints. Review and peer demonstration — Linear Algebra for Machine Learning: identify two likely mistakes, run a validation checklist, and explain improvements with evidence.
07Calculus & Optimization Basics18 hrs
Learn calculus & optimization basics through structured practical training at IDM Technologies, Vallioor. Cover derivatives, gradients, loss, and optimization concepts. Build a documented model experiment or AI workflow, practise evidence-based review, and demonstrate job-relevant skills. This is a draft module outline, subject to trainer approval.
Topics and subtopics
Derivatives: Derivatives: rate of change and slope; Derivatives: derivative rules and numerical approximation; Derivatives: graph a function and compare analytical and numerical change Gradients: gradients: partial derivatives and gradient direction; gradients: gradient descent step size and convergence; gradients: visualize optimization on a simple loss surface Loss: loss: loss functions and errors; loss: suitable metrics and optimization objectives; loss: compare predictions using a chosen loss Optimization concepts: optimization concepts: key definitions and purpose; optimization concepts: workflow and applied decisions in ai data science; optimization concepts: demonstration, error review and independent verification
Learning outcomes
Explain derivatives and distinguish correct use from a common misconception. Explain gradients and distinguish correct use from a common misconception. Explain loss and distinguish correct use from a common misconception. Explain optimization concepts and distinguish correct use from a common misconception. Apply Derivatives, Gradients, Loss to create a documented model experiment or AI workflow with independently checkable evidence. Evaluate the calculus & optimization basics deliverable against stated requirements, safety or quality criteria, and document corrections.
Practical activities
Workshop 1 — Derivatives: produce a focused example applying derivatives to a realistic brief; compare the initial result with a corrected result and record the reasoning. Submit: data provenance, evaluation criteria, limitations and reproducibility. Workshop 2 — Gradients: produce a focused example applying gradients to a realistic brief; compare the initial result with a corrected result and record the reasoning. Submit: data provenance, evaluation criteria, limitations and reproducibility. Workshop 3 — Loss: produce a focused example applying loss to a realistic brief; compare the initial result with a corrected result and record the reasoning. Submit: data provenance, evaluation criteria, limitations and reproducibility. Workshop 4 — Optimization concepts: produce a focused example applying optimization concepts to a realistic brief; compare the initial result with a corrected result and record the reasoning. Submit: data provenance, evaluation criteria, limitations and reproducibility. Applied mini-project — Calculus & Optimization Basics: assemble a documented model experiment or AI workflow combining Derivatives, Gradients, Loss; document inputs, outputs and constraints. Review and peer demonstration — Calculus & Optimization Basics: identify two likely mistakes, run a validation checklist, and explain improvements with evidence.
08NumPy & Pandas for Modelling22 hrs
Learn numpy & pandas for modelling through structured practical training at IDM Technologies, Vallioor. Cover arrays, preprocessing, feature tables. Build a documented model experiment or AI workflow, practise evidence-based review, and demonstrate job-relevant skills. This is a draft module outline, subject to trainer approval.
Topics and subtopics
Arrays: Arrays: key definitions and purpose; Arrays: workflow and applied decisions in ai data science; Arrays: demonstration, error review and independent verification Preprocessing: preprocessing: key definitions and purpose; preprocessing: workflow and applied decisions in ai data science; preprocessing: demonstration, error review and independent verification Feature tables: feature tables: key definitions and purpose; feature tables: workflow and applied decisions in ai data science; feature tables: demonstration, error review and independent verification
Learning outcomes
Explain arrays and distinguish correct use from a common misconception. Explain preprocessing and distinguish correct use from a common misconception. Explain feature tables and distinguish correct use from a common misconception. Apply Arrays, Preprocessing, Feature tables to create a documented model experiment or AI workflow with independently checkable evidence. Evaluate the numpy & pandas for modelling deliverable against stated requirements, safety or quality criteria, and document corrections.
Practical activities
Workshop 1 — Arrays: produce a focused example applying arrays to a realistic brief; compare the initial result with a corrected result and record the reasoning. Submit: data provenance, evaluation criteria, limitations and reproducibility. Workshop 2 — Preprocessing: produce a focused example applying preprocessing to a realistic brief; compare the initial result with a corrected result and record the reasoning. Submit: data provenance, evaluation criteria, limitations and reproducibility. Workshop 3 — Feature tables: produce a focused example applying feature tables to a realistic brief; compare the initial result with a corrected result and record the reasoning. Submit: data provenance, evaluation criteria, limitations and reproducibility. Applied mini-project — NumPy & Pandas for Modelling: assemble a documented model experiment or AI workflow combining Arrays, Preprocessing, Feature tables; document inputs, outputs and constraints. Review and peer demonstration — NumPy & Pandas for Modelling: identify two likely mistakes, run a validation checklist, and explain improvements with evidence.
09EDA & Visualization for Data Science18 hrs
Learn eda & visualization for data science through structured practical training at IDM Technologies, Vallioor. Cover patterns, distributions, leakage checks, and communication. Build a documented model experiment or AI workflow, practise evidence-based review, and demonstrate job-relevant skills. This is a draft module outline, subject to trainer approval.
Topics and subtopics
Patterns: Patterns: key definitions and purpose; Patterns: workflow and applied decisions in ai data science; Patterns: demonstration, error review and independent verification Distributions: distributions: key definitions and purpose; distributions: workflow and applied decisions in ai data science; distributions: demonstration, error review and independent verification Leakage checks: leakage checks: key definitions and purpose; leakage checks: workflow and applied decisions in ai data science; leakage checks: demonstration, error review and independent verification Communication: communication: key definitions and purpose; communication: workflow and applied decisions in ai data science; communication: demonstration, error review and independent verification
Learning outcomes
Explain patterns and distinguish correct use from a common misconception. Explain distributions and distinguish correct use from a common misconception. Explain leakage checks and distinguish correct use from a common misconception. Explain communication and distinguish correct use from a common misconception. Apply Patterns, Distributions, Leakage checks to create a documented model experiment or AI workflow with independently checkable evidence. Evaluate the eda & visualization for data science deliverable against stated requirements, safety or quality criteria, and document corrections.
Practical activities
Workshop 1 — Patterns: produce a focused example applying patterns to a realistic brief; compare the initial result with a corrected result and record the reasoning. Submit: data provenance, evaluation criteria, limitations and reproducibility. Workshop 2 — Distributions: produce a focused example applying distributions to a realistic brief; compare the initial result with a corrected result and record the reasoning. Submit: data provenance, evaluation criteria, limitations and reproducibility. Workshop 3 — Leakage checks: produce a focused example applying leakage checks to a realistic brief; compare the initial result with a corrected result and record the reasoning. Submit: data provenance, evaluation criteria, limitations and reproducibility. Workshop 4 — Communication: produce a focused example applying communication to a realistic brief; compare the initial result with a corrected result and record the reasoning. Submit: data provenance, evaluation criteria, limitations and reproducibility. Applied mini-project — EDA & Visualization for Data Science: assemble a documented model experiment or AI workflow combining Patterns, Distributions, Leakage checks; document inputs, outputs and constraints. Review and peer demonstration — EDA & Visualization for Data Science: identify two likely mistakes, run a validation checklist, and explain improvements with evidence.
10AI Foundations & Responsible AI16 hrs
Learn ai foundations & responsible ai through structured practical training at IDM Technologies, Vallioor. Cover ml concepts, use cases, bias, and data rights. Build a documented model experiment or AI workflow, practise evidence-based review, and demonstrate job-relevant skills. This is a draft module outline, subject to trainer approval.
Topics and subtopics
ML concepts: ML concepts: key definitions and purpose; ML concepts: workflow and applied decisions in ai data science; ML concepts: demonstration, error review and independent verification Use cases: use cases: key definitions and purpose; use cases: workflow and applied decisions in ai data science; use cases: demonstration, error review and independent verification Bias: bias: key definitions and purpose; bias: workflow and applied decisions in ai data science; bias: demonstration, error review and independent verification Data rights: data rights: key definitions and purpose; data rights: workflow and applied decisions in ai data science; data rights: demonstration, error review and independent verification Evaluation limits: evaluation limits: key definitions and purpose; evaluation limits: workflow and applied decisions in ai data science; evaluation limits: demonstration, error review and independent verification
Learning outcomes
Explain ml concepts and distinguish correct use from a common misconception. Explain use cases and distinguish correct use from a common misconception. Explain bias and distinguish correct use from a common misconception. Explain data rights and distinguish correct use from a common misconception. Explain evaluation limits and distinguish correct use from a common misconception. Apply ML concepts, Use cases, Bias to create a documented model experiment or AI workflow with independently checkable evidence. Evaluate the ai foundations & responsible ai deliverable against stated requirements, safety or quality criteria, and document corrections.
Practical activities
Workshop 1 — ML concepts: produce a focused example applying ml concepts to a realistic brief; compare the initial result with a corrected result and record the reasoning. Submit: data provenance, evaluation criteria, limitations and reproducibility. Workshop 2 — Use cases: produce a focused example applying use cases to a realistic brief; compare the initial result with a corrected result and record the reasoning. Submit: data provenance, evaluation criteria, limitations and reproducibility. Workshop 3 — Bias: produce a focused example applying bias to a realistic brief; compare the initial result with a corrected result and record the reasoning. Submit: data provenance, evaluation criteria, limitations and reproducibility. Workshop 4 — Data rights: produce a focused example applying data rights to a realistic brief; compare the initial result with a corrected result and record the reasoning. Submit: data provenance, evaluation criteria, limitations and reproducibility. Workshop 5 — Evaluation limits: produce a focused example applying evaluation limits to a realistic brief; compare the initial result with a corrected result and record the reasoning. Submit: data provenance, evaluation criteria, limitations and reproducibility. Applied mini-project — AI Foundations & Responsible AI: assemble a documented model experiment or AI workflow combining ML concepts, Use cases, Bias; document inputs, outputs and constraints. Review and peer demonstration — AI Foundations & Responsible AI: identify two likely mistakes, run a validation checklist, and explain improvements with evidence.
11Supervised Learning: Regression22 hrs
Learn supervised learning: regression through structured practical training at IDM Technologies, Vallioor. Cover linear models, regularization, metrics, and pipelines. Build a documented model experiment or AI workflow, practise evidence-based review, and demonstrate job-relevant skills. This is a draft module outline, subject to trainer approval.
Topics and subtopics
Linear models: Linear models: inputs, assumptions and baseline selection; Linear models: evaluation, errors and leakage risks; Linear models: record a reproducible comparison experiment Regularization: regularization: key definitions and purpose; regularization: workflow and applied decisions in ai data science; regularization: demonstration, error review and independent verification Metrics: metrics: key definitions and purpose; metrics: workflow and applied decisions in ai data science; metrics: demonstration, error review and independent verification Pipelines: pipelines: key definitions and purpose; pipelines: workflow and applied decisions in ai data science; pipelines: demonstration, error review and independent verification
Learning outcomes
Explain linear models and distinguish correct use from a common misconception. Explain regularization and distinguish correct use from a common misconception. Explain metrics and distinguish correct use from a common misconception. Explain pipelines and distinguish correct use from a common misconception. Apply Linear models, Regularization, Metrics to create a documented model experiment or AI workflow with independently checkable evidence. Evaluate the supervised learning: regression deliverable against stated requirements, safety or quality criteria, and document corrections.
Practical activities
Workshop 1 — Linear models: produce a focused example applying linear models to a realistic brief; compare the initial result with a corrected result and record the reasoning. Submit: data provenance, evaluation criteria, limitations and reproducibility. Workshop 2 — Regularization: produce a focused example applying regularization to a realistic brief; compare the initial result with a corrected result and record the reasoning. Submit: data provenance, evaluation criteria, limitations and reproducibility. Workshop 3 — Metrics: produce a focused example applying metrics to a realistic brief; compare the initial result with a corrected result and record the reasoning. Submit: data provenance, evaluation criteria, limitations and reproducibility. Workshop 4 — Pipelines: produce a focused example applying pipelines to a realistic brief; compare the initial result with a corrected result and record the reasoning. Submit: data provenance, evaluation criteria, limitations and reproducibility. Applied mini-project — Supervised Learning: Regression: assemble a documented model experiment or AI workflow combining Linear models, Regularization, Metrics; document inputs, outputs and constraints. Review and peer demonstration — Supervised Learning: Regression: identify two likely mistakes, run a validation checklist, and explain improvements with evidence.
12Supervised Learning: Classification24 hrs
Learn supervised learning: classification through structured practical training at IDM Technologies, Vallioor. Cover logistic, trees, ensembles, and confusion matrix. Build a documented model experiment or AI workflow, practise evidence-based review, and demonstrate job-relevant skills. This is a draft module outline, subject to trainer approval.
Topics and subtopics
Logistic: Logistic: key definitions and purpose; Logistic: workflow and applied decisions in ai data science; Logistic: demonstration, error review and independent verification Trees: trees: key definitions and purpose; trees: workflow and applied decisions in ai data science; trees: demonstration, error review and independent verification Ensembles: ensembles: key definitions and purpose; ensembles: workflow and applied decisions in ai data science; ensembles: demonstration, error review and independent verification Confusion matrix: confusion matrix: key definitions and purpose; confusion matrix: workflow and applied decisions in ai data science; confusion matrix: demonstration, error review and independent verification
Learning outcomes
Explain logistic and distinguish correct use from a common misconception. Explain trees and distinguish correct use from a common misconception. Explain ensembles and distinguish correct use from a common misconception. Explain confusion matrix and distinguish correct use from a common misconception. Apply Logistic, Trees, Ensembles to create a documented model experiment or AI workflow with independently checkable evidence. Evaluate the supervised learning: classification deliverable against stated requirements, safety or quality criteria, and document corrections.
Practical activities
Workshop 1 — Logistic: produce a focused example applying logistic to a realistic brief; compare the initial result with a corrected result and record the reasoning. Submit: data provenance, evaluation criteria, limitations and reproducibility. Workshop 2 — Trees: produce a focused example applying trees to a realistic brief; compare the initial result with a corrected result and record the reasoning. Submit: data provenance, evaluation criteria, limitations and reproducibility. Workshop 3 — Ensembles: produce a focused example applying ensembles to a realistic brief; compare the initial result with a corrected result and record the reasoning. Submit: data provenance, evaluation criteria, limitations and reproducibility. Workshop 4 — Confusion matrix: produce a focused example applying confusion matrix to a realistic brief; compare the initial result with a corrected result and record the reasoning. Submit: data provenance, evaluation criteria, limitations and reproducibility. Applied mini-project — Supervised Learning: Classification: assemble a documented model experiment or AI workflow combining Logistic, Trees, Ensembles; document inputs, outputs and constraints. Review and peer demonstration — Supervised Learning: Classification: identify two likely mistakes, run a validation checklist, and explain improvements with evidence.
13Unsupervised Learning20 hrs
Learn unsupervised learning through structured practical training at IDM Technologies, Vallioor. Cover clustering, pca, anomaly detection. Build a documented model experiment or AI workflow, practise evidence-based review, and demonstrate job-relevant skills. This is a draft module outline, subject to trainer approval.
Topics and subtopics
Clustering: Clustering: key definitions and purpose; Clustering: workflow and applied decisions in ai data science; Clustering: demonstration, error review and independent verification PCA: PCA: key definitions and purpose; PCA: workflow and applied decisions in ai data science; PCA: demonstration, error review and independent verification Anomaly detection: anomaly detection: key definitions and purpose; anomaly detection: workflow and applied decisions in ai data science; anomaly detection: demonstration, error review and independent verification
Learning outcomes
Explain clustering and distinguish correct use from a common misconception. Explain pca and distinguish correct use from a common misconception. Explain anomaly detection and distinguish correct use from a common misconception. Apply Clustering, PCA, Anomaly detection to create a documented model experiment or AI workflow with independently checkable evidence. Evaluate the unsupervised learning deliverable against stated requirements, safety or quality criteria, and document corrections.
Practical activities
Workshop 1 — Clustering: produce a focused example applying clustering to a realistic brief; compare the initial result with a corrected result and record the reasoning. Submit: data provenance, evaluation criteria, limitations and reproducibility. Workshop 2 — PCA: produce a focused example applying pca to a realistic brief; compare the initial result with a corrected result and record the reasoning. Submit: data provenance, evaluation criteria, limitations and reproducibility. Workshop 3 — Anomaly detection: produce a focused example applying anomaly detection to a realistic brief; compare the initial result with a corrected result and record the reasoning. Submit: data provenance, evaluation criteria, limitations and reproducibility. Applied mini-project — Unsupervised Learning: assemble a documented model experiment or AI workflow combining Clustering, PCA, Anomaly detection; document inputs, outputs and constraints. Review and peer demonstration — Unsupervised Learning: identify two likely mistakes, run a validation checklist, and explain improvements with evidence.
14Feature Engineering & Model Evaluation22 hrs
Learn feature engineering & model evaluation through structured practical training at IDM Technologies, Vallioor. Cover cross-validation, data leakage, drift, and selection. Build a documented model experiment or AI workflow, practise evidence-based review, and demonstrate job-relevant skills. This is a draft module outline, subject to trainer approval.
Topics and subtopics
Cross-validation: Cross-validation: key definitions and purpose; Cross-validation: workflow and applied decisions in ai data science; Cross-validation: demonstration, error review and independent verification Data leakage: data leakage: key definitions and purpose; data leakage: workflow and applied decisions in ai data science; data leakage: demonstration, error review and independent verification Drift: drift: key definitions and purpose; drift: workflow and applied decisions in ai data science; drift: demonstration, error review and independent verification Selection: selection: key definitions and purpose; selection: workflow and applied decisions in ai data science; selection: demonstration, error review and independent verification
Learning outcomes
Explain cross-validation and distinguish correct use from a common misconception. Explain data leakage and distinguish correct use from a common misconception. Explain drift and distinguish correct use from a common misconception. Explain selection and distinguish correct use from a common misconception. Apply Cross-validation, Data leakage, Drift to create a documented model experiment or AI workflow with independently checkable evidence. Evaluate the feature engineering & model evaluation deliverable against stated requirements, safety or quality criteria, and document corrections.
Practical activities
Workshop 1 — Cross-validation: produce a focused example applying cross-validation to a realistic brief; compare the initial result with a corrected result and record the reasoning. Submit: data provenance, evaluation criteria, limitations and reproducibility. Workshop 2 — Data leakage: produce a focused example applying data leakage to a realistic brief; compare the initial result with a corrected result and record the reasoning. Submit: data provenance, evaluation criteria, limitations and reproducibility. Workshop 3 — Drift: produce a focused example applying drift to a realistic brief; compare the initial result with a corrected result and record the reasoning. Submit: data provenance, evaluation criteria, limitations and reproducibility. Workshop 4 — Selection: produce a focused example applying selection to a realistic brief; compare the initial result with a corrected result and record the reasoning. Submit: data provenance, evaluation criteria, limitations and reproducibility. Applied mini-project — Feature Engineering & Model Evaluation: assemble a documented model experiment or AI workflow combining Cross-validation, Data leakage, Drift; document inputs, outputs and constraints. Review and peer demonstration — Feature Engineering & Model Evaluation: identify two likely mistakes, run a validation checklist, and explain improvements with evidence.
15Time Series & Forecasting20 hrs
Learn time series & forecasting through structured practical training at IDM Technologies, Vallioor. Cover trend, seasonality, lags, and backtesting. Build a documented model experiment or AI workflow, practise evidence-based review, and demonstrate job-relevant skills. This is a draft module outline, subject to trainer approval.
Topics and subtopics
Trend: Trend: key definitions and purpose; Trend: workflow and applied decisions in ai data science; Trend: demonstration, error review and independent verification Seasonality: seasonality: key definitions and purpose; seasonality: workflow and applied decisions in ai data science; seasonality: demonstration, error review and independent verification Lags: lags: key definitions and purpose; lags: workflow and applied decisions in ai data science; lags: demonstration, error review and independent verification Backtesting: backtesting: test cases, expected results and fixtures; backtesting: negative cases, regression and bug reports; backtesting: write and run repeatable validation checks
Learning outcomes
Explain trend and distinguish correct use from a common misconception. Explain seasonality and distinguish correct use from a common misconception. Explain lags and distinguish correct use from a common misconception. Explain backtesting and distinguish correct use from a common misconception. Apply Trend, Seasonality, Lags to create a documented model experiment or AI workflow with independently checkable evidence. Evaluate the time series & forecasting deliverable against stated requirements, safety or quality criteria, and document corrections.
Practical activities
Workshop 1 — Trend: produce a focused example applying trend to a realistic brief; compare the initial result with a corrected result and record the reasoning. Submit: data provenance, evaluation criteria, limitations and reproducibility. Workshop 2 — Seasonality: produce a focused example applying seasonality to a realistic brief; compare the initial result with a corrected result and record the reasoning. Submit: data provenance, evaluation criteria, limitations and reproducibility. Workshop 3 — Lags: produce a focused example applying lags to a realistic brief; compare the initial result with a corrected result and record the reasoning. Submit: data provenance, evaluation criteria, limitations and reproducibility. Workshop 4 — Backtesting: produce a focused example applying backtesting to a realistic brief; compare the initial result with a corrected result and record the reasoning. Submit: data provenance, evaluation criteria, limitations and reproducibility. Applied mini-project — Time Series & Forecasting: assemble a documented model experiment or AI workflow combining Trend, Seasonality, Lags; document inputs, outputs and constraints. Review and peer demonstration — Time Series & Forecasting: identify two likely mistakes, run a validation checklist, and explain improvements with evidence.
16Neural Network Foundations24 hrs
Learn neural network foundations through structured practical training at IDM Technologies, Vallioor. Cover perceptrons, backpropagation, frameworks. Build a documented model experiment or AI workflow, practise evidence-based review, and demonstrate job-relevant skills. This is a draft module outline, subject to trainer approval.
Topics and subtopics
Perceptrons: Perceptrons: key definitions and purpose; Perceptrons: workflow and applied decisions in ai data science; Perceptrons: demonstration, error review and independent verification Backpropagation: backpropagation: key definitions and purpose; backpropagation: workflow and applied decisions in ai data science; backpropagation: demonstration, error review and independent verification Frameworks: frameworks: key definitions and purpose; frameworks: workflow and applied decisions in ai data science; frameworks: demonstration, error review and independent verification
Learning outcomes
Explain perceptrons and distinguish correct use from a common misconception. Explain backpropagation and distinguish correct use from a common misconception. Explain frameworks and distinguish correct use from a common misconception. Apply Perceptrons, Backpropagation, Frameworks to create a documented model experiment or AI workflow with independently checkable evidence. Evaluate the neural network foundations deliverable against stated requirements, safety or quality criteria, and document corrections.
Practical activities
Workshop 1 — Perceptrons: produce a focused example applying perceptrons to a realistic brief; compare the initial result with a corrected result and record the reasoning. Submit: data provenance, evaluation criteria, limitations and reproducibility. Workshop 2 — Backpropagation: produce a focused example applying backpropagation to a realistic brief; compare the initial result with a corrected result and record the reasoning. Submit: data provenance, evaluation criteria, limitations and reproducibility. Workshop 3 — Frameworks: produce a focused example applying frameworks to a realistic brief; compare the initial result with a corrected result and record the reasoning. Submit: data provenance, evaluation criteria, limitations and reproducibility. Applied mini-project — Neural Network Foundations: assemble a documented model experiment or AI workflow combining Perceptrons, Backpropagation, Frameworks; document inputs, outputs and constraints. Review and peer demonstration — Neural Network Foundations: identify two likely mistakes, run a validation checklist, and explain improvements with evidence.
17Deep Learning for Images & Text22 hrs
Learn deep learning for images & text through structured practical training at IDM Technologies, Vallioor. Cover cnn, embeddings, transfer learning, and evaluation. Build a documented model experiment or AI workflow, practise evidence-based review, and demonstrate job-relevant skills. This is a draft module outline, subject to trainer approval.
Topics and subtopics
CNN: CNN: key definitions and purpose; CNN: workflow and applied decisions in ai data science; CNN: demonstration, error review and independent verification Embeddings: embeddings: key definitions and purpose; embeddings: workflow and applied decisions in ai data science; embeddings: demonstration, error review and independent verification Transfer learning: transfer learning: key definitions and purpose; transfer learning: workflow and applied decisions in ai data science; transfer learning: demonstration, error review and independent verification Evaluation: evaluation: key definitions and purpose; evaluation: workflow and applied decisions in ai data science; evaluation: demonstration, error review and independent verification
Learning outcomes
Explain cnn and distinguish correct use from a common misconception. Explain embeddings and distinguish correct use from a common misconception. Explain transfer learning and distinguish correct use from a common misconception. Explain evaluation and distinguish correct use from a common misconception. Apply CNN, Embeddings, Transfer learning to create a documented model experiment or AI workflow with independently checkable evidence. Evaluate the deep learning for images & text deliverable against stated requirements, safety or quality criteria, and document corrections.
Practical activities
Workshop 1 — CNN: produce a focused example applying cnn to a realistic brief; compare the initial result with a corrected result and record the reasoning. Submit: data provenance, evaluation criteria, limitations and reproducibility. Workshop 2 — Embeddings: produce a focused example applying embeddings to a realistic brief; compare the initial result with a corrected result and record the reasoning. Submit: data provenance, evaluation criteria, limitations and reproducibility. Workshop 3 — Transfer learning: produce a focused example applying transfer learning to a realistic brief; compare the initial result with a corrected result and record the reasoning. Submit: data provenance, evaluation criteria, limitations and reproducibility. Workshop 4 — Evaluation: produce a focused example applying evaluation to a realistic brief; compare the initial result with a corrected result and record the reasoning. Submit: data provenance, evaluation criteria, limitations and reproducibility. Applied mini-project — Deep Learning for Images & Text: assemble a documented model experiment or AI workflow combining CNN, Embeddings, Transfer learning; document inputs, outputs and constraints. Review and peer demonstration — Deep Learning for Images & Text: identify two likely mistakes, run a validation checklist, and explain improvements with evidence.
18NLP Foundations20 hrs
Learn nlp foundations through structured practical training at IDM Technologies, Vallioor. Cover tokenization, classification, similarity, and embeddings. Build a documented model experiment or AI workflow, practise evidence-based review, and demonstrate job-relevant skills. This is a draft module outline, subject to trainer approval.
Topics and subtopics
Tokenization: Tokenization: key definitions and purpose; Tokenization: workflow and applied decisions in ai data science; Tokenization: demonstration, error review and independent verification Classification: classification: key definitions and purpose; classification: workflow and applied decisions in ai data science; classification: demonstration, error review and independent verification Similarity: similarity: key definitions and purpose; similarity: workflow and applied decisions in ai data science; similarity: demonstration, error review and independent verification Embeddings: embeddings: key definitions and purpose; embeddings: workflow and applied decisions in ai data science; embeddings: demonstration, error review and independent verification
Learning outcomes
Explain tokenization and distinguish correct use from a common misconception. Explain classification and distinguish correct use from a common misconception. Explain similarity and distinguish correct use from a common misconception. Explain embeddings and distinguish correct use from a common misconception. Apply Tokenization, Classification, Similarity to create a documented model experiment or AI workflow with independently checkable evidence. Evaluate the nlp foundations deliverable against stated requirements, safety or quality criteria, and document corrections.
Practical activities
Workshop 1 — Tokenization: produce a focused example applying tokenization to a realistic brief; compare the initial result with a corrected result and record the reasoning. Submit: data provenance, evaluation criteria, limitations and reproducibility. Workshop 2 — Classification: produce a focused example applying classification to a realistic brief; compare the initial result with a corrected result and record the reasoning. Submit: data provenance, evaluation criteria, limitations and reproducibility. Workshop 3 — Similarity: produce a focused example applying similarity to a realistic brief; compare the initial result with a corrected result and record the reasoning. Submit: data provenance, evaluation criteria, limitations and reproducibility. Workshop 4 — Embeddings: produce a focused example applying embeddings to a realistic brief; compare the initial result with a corrected result and record the reasoning. Submit: data provenance, evaluation criteria, limitations and reproducibility. Applied mini-project — NLP Foundations: assemble a documented model experiment or AI workflow combining Tokenization, Classification, Similarity; document inputs, outputs and constraints. Review and peer demonstration — NLP Foundations: identify two likely mistakes, run a validation checklist, and explain improvements with evidence.
19Generative AI & LLM Application Design24 hrs
Learn generative ai & llm application design through structured practical training at IDM Technologies, Vallioor. Cover prompts, rag, evaluation, and hallucination controls. Build a documented model experiment or AI workflow, practise evidence-based review, and demonstrate job-relevant skills. This is a draft module outline, subject to trainer approval.
Topics and subtopics
Prompts: Prompts: key definitions and purpose; Prompts: workflow and applied decisions in ai data science; Prompts: demonstration, error review and independent verification RAG: RAG: key definitions and purpose; RAG: workflow and applied decisions in ai data science; RAG: demonstration, error review and independent verification Evaluation: evaluation: key definitions and purpose; evaluation: workflow and applied decisions in ai data science; evaluation: demonstration, error review and independent verification Hallucination controls: hallucination controls: key definitions and purpose; hallucination controls: workflow and applied decisions in ai data science; hallucination controls: demonstration, error review and independent verification
Learning outcomes
Explain prompts and distinguish correct use from a common misconception. Explain rag and distinguish correct use from a common misconception. Explain evaluation and distinguish correct use from a common misconception. Explain hallucination controls and distinguish correct use from a common misconception. Apply Prompts, RAG, Evaluation to create a documented model experiment or AI workflow with independently checkable evidence. Evaluate the generative ai & llm application design deliverable against stated requirements, safety or quality criteria, and document corrections.
Practical activities
Workshop 1 — Prompts: produce a focused example applying prompts to a realistic brief; compare the initial result with a corrected result and record the reasoning. Submit: data provenance, evaluation criteria, limitations and reproducibility. Workshop 2 — RAG: produce a focused example applying rag to a realistic brief; compare the initial result with a corrected result and record the reasoning. Submit: data provenance, evaluation criteria, limitations and reproducibility. Workshop 3 — Evaluation: produce a focused example applying evaluation to a realistic brief; compare the initial result with a corrected result and record the reasoning. Submit: data provenance, evaluation criteria, limitations and reproducibility. Workshop 4 — Hallucination controls: produce a focused example applying hallucination controls to a realistic brief; compare the initial result with a corrected result and record the reasoning. Submit: data provenance, evaluation criteria, limitations and reproducibility. Applied mini-project — Generative AI & LLM Application Design: assemble a documented model experiment or AI workflow combining Prompts, RAG, Evaluation; document inputs, outputs and constraints. Review and peer demonstration — Generative AI & LLM Application Design: identify two likely mistakes, run a validation checklist, and explain improvements with evidence.
20Model Deployment & MLOps Basics22 hrs
Learn model deployment & mlops basics through structured practical training at IDM Technologies, Vallioor. Cover inference apis, containers, model versioning, and monitoring. Build a documented model experiment or AI workflow, practise evidence-based review, and demonstrate job-relevant skills. This is a draft module outline, subject to trainer approval.
Topics and subtopics
Inference APIs: Inference APIs: HTTP verbs, request bodies and response codes; Inference APIs: validation, error handling and access rules; Inference APIs: exercise endpoints with positive and negative test cases Containers: containers: key definitions and purpose; containers: workflow and applied decisions in ai data science; containers: demonstration, error review and independent verification Model versioning: model versioning: inputs, assumptions and baseline selection; model versioning: evaluation, errors and leakage risks; model versioning: record a reproducible comparison experiment Monitoring: monitoring: key definitions and purpose; monitoring: workflow and applied decisions in ai data science; monitoring: demonstration, error review and independent verification
Learning outcomes
Explain inference apis and distinguish correct use from a common misconception. Explain containers and distinguish correct use from a common misconception. Explain model versioning and distinguish correct use from a common misconception. Explain monitoring and distinguish correct use from a common misconception. Apply Inference APIs, Containers, Model versioning to create a documented model experiment or AI workflow with independently checkable evidence. Evaluate the model deployment & mlops basics deliverable against stated requirements, safety or quality criteria, and document corrections.
Practical activities
Workshop 1 — Inference APIs: produce a focused example applying inference apis to a realistic brief; compare the initial result with a corrected result and record the reasoning. Submit: data provenance, evaluation criteria, limitations and reproducibility. Workshop 2 — Containers: produce a focused example applying containers to a realistic brief; compare the initial result with a corrected result and record the reasoning. Submit: data provenance, evaluation criteria, limitations and reproducibility. Workshop 3 — Model versioning: produce a focused example applying model versioning to a realistic brief; compare the initial result with a corrected result and record the reasoning. Submit: data provenance, evaluation criteria, limitations and reproducibility. Workshop 4 — Monitoring: produce a focused example applying monitoring to a realistic brief; compare the initial result with a corrected result and record the reasoning. Submit: data provenance, evaluation criteria, limitations and reproducibility. Applied mini-project — Model Deployment & MLOps Basics: assemble a documented model experiment or AI workflow combining Inference APIs, Containers, Model versioning; document inputs, outputs and constraints. Review and peer demonstration — Model Deployment & MLOps Basics: identify two likely mistakes, run a validation checklist, and explain improvements with evidence.
21Responsible AI & Model Governance18 hrs
Learn responsible ai & model governance through structured practical training at IDM Technologies, Vallioor. Cover privacy, fairness, transparency, and safety evaluations. Build a documented model experiment or AI workflow, practise evidence-based review, and demonstrate job-relevant skills. This is a draft module outline, subject to trainer approval.
Topics and subtopics
Privacy: Privacy: key definitions and purpose; Privacy: workflow and applied decisions in ai data science; Privacy: demonstration, error review and independent verification Fairness: fairness: key definitions and purpose; fairness: workflow and applied decisions in ai data science; fairness: demonstration, error review and independent verification Transparency: transparency: key definitions and purpose; transparency: workflow and applied decisions in ai data science; transparency: demonstration, error review and independent verification Safety evaluations: safety evaluations: key definitions and purpose; safety evaluations: workflow and applied decisions in ai data science; safety evaluations: demonstration, error review and independent verification
Learning outcomes
Explain privacy and distinguish correct use from a common misconception. Explain fairness and distinguish correct use from a common misconception. Explain transparency and distinguish correct use from a common misconception. Explain safety evaluations and distinguish correct use from a common misconception. Apply Privacy, Fairness, Transparency to create a documented model experiment or AI workflow with independently checkable evidence. Evaluate the responsible ai & model governance deliverable against stated requirements, safety or quality criteria, and document corrections.
Practical activities
Workshop 1 — Privacy: produce a focused example applying privacy to a realistic brief; compare the initial result with a corrected result and record the reasoning. Submit: data provenance, evaluation criteria, limitations and reproducibility. Workshop 2 — Fairness: produce a focused example applying fairness to a realistic brief; compare the initial result with a corrected result and record the reasoning. Submit: data provenance, evaluation criteria, limitations and reproducibility. Workshop 3 — Transparency: produce a focused example applying transparency to a realistic brief; compare the initial result with a corrected result and record the reasoning. Submit: data provenance, evaluation criteria, limitations and reproducibility. Workshop 4 — Safety evaluations: produce a focused example applying safety evaluations to a realistic brief; compare the initial result with a corrected result and record the reasoning. Submit: data provenance, evaluation criteria, limitations and reproducibility. Applied mini-project — Responsible AI & Model Governance: assemble a documented model experiment or AI workflow combining Privacy, Fairness, Transparency; document inputs, outputs and constraints. Review and peer demonstration — Responsible AI & Model Governance: identify two likely mistakes, run a validation checklist, and explain improvements with evidence.
22Experiment Design & Reproducibility16 hrs
Learn experiment design & reproducibility through structured practical training at IDM Technologies, Vallioor. Cover baselines, logging, reproducibility, and documentation. Build a documented model experiment or AI workflow, practise evidence-based review, and demonstrate job-relevant skills. This is a draft module outline, subject to trainer approval.
Topics and subtopics
Baselines: Baselines: key definitions and purpose; Baselines: workflow and applied decisions in ai data science; Baselines: demonstration, error review and independent verification Logging: logging: key definitions and purpose; logging: workflow and applied decisions in ai data science; logging: demonstration, error review and independent verification Reproducibility: reproducibility: key definitions and purpose; reproducibility: workflow and applied decisions in ai data science; reproducibility: demonstration, error review and independent verification Documentation: documentation: key definitions and purpose; documentation: workflow and applied decisions in ai data science; documentation: demonstration, error review and independent verification
Learning outcomes
Explain baselines and distinguish correct use from a common misconception. Explain logging and distinguish correct use from a common misconception. Explain reproducibility and distinguish correct use from a common misconception. Explain documentation and distinguish correct use from a common misconception. Apply Baselines, Logging, Reproducibility to create a documented model experiment or AI workflow with independently checkable evidence. Evaluate the experiment design & reproducibility deliverable against stated requirements, safety or quality criteria, and document corrections.
Practical activities
Workshop 1 — Baselines: produce a focused example applying baselines to a realistic brief; compare the initial result with a corrected result and record the reasoning. Submit: data provenance, evaluation criteria, limitations and reproducibility. Workshop 2 — Logging: produce a focused example applying logging to a realistic brief; compare the initial result with a corrected result and record the reasoning. Submit: data provenance, evaluation criteria, limitations and reproducibility. Workshop 3 — Reproducibility: produce a focused example applying reproducibility to a realistic brief; compare the initial result with a corrected result and record the reasoning. Submit: data provenance, evaluation criteria, limitations and reproducibility. Workshop 4 — Documentation: produce a focused example applying documentation to a realistic brief; compare the initial result with a corrected result and record the reasoning. Submit: data provenance, evaluation criteria, limitations and reproducibility. Applied mini-project — Experiment Design & Reproducibility: assemble a documented model experiment or AI workflow combining Baselines, Logging, Reproducibility; document inputs, outputs and constraints. Review and peer demonstration — Experiment Design & Reproducibility: identify two likely mistakes, run a validation checklist, and explain improvements with evidence.
23AI & Data Science Capstone40 hrs
Learn ai & data science capstone through structured practical training at IDM Technologies, Vallioor. Cover problem definition, dataset, training, and validation. Build a documented model experiment or AI workflow, practise evidence-based review, and demonstrate job-relevant skills. This is a draft module outline, subject to trainer approval.
Topics and subtopics
Problem definition: Problem definition: key definitions and purpose; Problem definition: workflow and applied decisions in ai data science; Problem definition: demonstration, error review and independent verification Dataset: dataset: key definitions and purpose; dataset: workflow and applied decisions in ai data science; dataset: demonstration, error review and independent verification Training: training: key definitions and purpose; training: workflow and applied decisions in ai data science; training: demonstration, error review and independent verification Validation: validation: key definitions and purpose; validation: workflow and applied decisions in ai data science; validation: demonstration, error review and independent verification Demo: demo: key definitions and purpose; demo: workflow and applied decisions in ai data science; demo: demonstration, error review and independent verification
Learning outcomes
Explain problem definition and distinguish correct use from a common misconception. Explain dataset and distinguish correct use from a common misconception. Explain training and distinguish correct use from a common misconception. Explain validation and distinguish correct use from a common misconception. Explain demo and distinguish correct use from a common misconception. Apply Problem definition, Dataset, Training to create a documented model experiment or AI workflow with independently checkable evidence. Evaluate the ai & data science capstone deliverable against stated requirements, safety or quality criteria, and document corrections.
Practical activities
Workshop 1 — Problem definition: produce a focused example applying problem definition to a realistic brief; compare the initial result with a corrected result and record the reasoning. Submit: data provenance, evaluation criteria, limitations and reproducibility. Workshop 2 — Dataset: produce a focused example applying dataset to a realistic brief; compare the initial result with a corrected result and record the reasoning. Submit: data provenance, evaluation criteria, limitations and reproducibility. Workshop 3 — Training: produce a focused example applying training to a realistic brief; compare the initial result with a corrected result and record the reasoning. Submit: data provenance, evaluation criteria, limitations and reproducibility. Workshop 4 — Validation: produce a focused example applying validation to a realistic brief; compare the initial result with a corrected result and record the reasoning. Submit: data provenance, evaluation criteria, limitations and reproducibility. Workshop 5 — Demo: produce a focused example applying demo to a realistic brief; compare the initial result with a corrected result and record the reasoning. Submit: data provenance, evaluation criteria, limitations and reproducibility. Applied mini-project — AI & Data Science Capstone: assemble a documented model experiment or AI workflow combining Problem definition, Dataset, Training; document inputs, outputs and constraints. Review and peer demonstration — AI & Data Science Capstone: identify two likely mistakes, run a validation checklist, and explain improvements with evidence.
24Project Planning, Documentation & Presentation12 hrs
Learn project planning, documentation & presentation through structured practical training at IDM Technologies, Vallioor. Cover requirements, milestones, documentation, and demo. Build a working software feature or technical demonstration, practise evidence-based review, and demonstrate job-relevant skills. This is a draft module outline, subject to trainer approval.
Topics and subtopics
Requirements: Requirements: key definitions and purpose; Requirements: workflow and applied decisions in software development; Requirements: demonstration, error review and independent verification Milestones: milestones: key definitions and purpose; milestones: workflow and applied decisions in software development; milestones: demonstration, error review and independent verification Documentation: documentation: key definitions and purpose; documentation: workflow and applied decisions in software development; documentation: demonstration, error review and independent verification Demo: demo: key definitions and purpose; demo: workflow and applied decisions in software development; demo: demonstration, error review and independent verification Presentation: presentation: key definitions and purpose; presentation: workflow and applied decisions in software development; presentation: demonstration, error review and independent verification
Learning outcomes
Explain requirements and distinguish correct use from a common misconception. Explain milestones and distinguish correct use from a common misconception. Explain documentation and distinguish correct use from a common misconception. Explain demo and distinguish correct use from a common misconception. Explain presentation and distinguish correct use from a common misconception. Apply Requirements, Milestones, Documentation to create a working software feature or technical demonstration with independently checkable evidence. Evaluate the project planning, documentation & presentation deliverable against stated requirements, safety or quality criteria, and document corrections.
Practical activities
Workshop 1 — Requirements: produce a focused example applying requirements to a realistic brief; compare the initial result with a corrected result and record the reasoning. Submit: source code, tests, setup notes and a short walkthrough. Workshop 2 — Milestones: produce a focused example applying milestones to a realistic brief; compare the initial result with a corrected result and record the reasoning. Submit: source code, tests, setup notes and a short walkthrough. Workshop 3 — Documentation: produce a focused example applying documentation to a realistic brief; compare the initial result with a corrected result and record the reasoning. Submit: source code, tests, setup notes and a short walkthrough. Workshop 4 — Demo: produce a focused example applying demo to a realistic brief; compare the initial result with a corrected result and record the reasoning. Submit: source code, tests, setup notes and a short walkthrough. Workshop 5 — Presentation: produce a focused example applying presentation to a realistic brief; compare the initial result with a corrected result and record the reasoning. Submit: source code, tests, setup notes and a short walkthrough. Applied mini-project — Project Planning, Documentation & Presentation: assemble a working software feature or technical demonstration combining Requirements, Milestones, Documentation; document inputs, outputs and constraints. Review and peer demonstration — Project Planning, Documentation & Presentation: identify two likely mistakes, run a validation checklist, and explain improvements with evidence.