Become a certified Data Scientist or Generative AI Engineer.
Two ~3-month professional programmes, each with hands-on JupyterLab practice in every section, a real-world case study in every section and a portfolio capstone. Nine expert-written sections take you from the foundations to production — with a clear route to Vecorra certification.
Two routes. One rigorous standard.
Pick the programme that matches where you want to work — both are built around the skills employers actually assess, and both lead to a Vecorra certification.
Become a certified Data Scientist.
A ~3-month professional programme that takes you from Python, SQL and statistics through machine learning, model evaluation, responsible AI, deployment and a portfolio capstone — with hands-on JupyterLab practice in every section and a route to Vecorra certification.
9 sections
- Python, SQL and data analysis foundations
- Statistics: probability, hypothesis testing, confidence intervals
- Machine learning modelling with scikit-learn
- Model evaluation, cross-validation and leakage-avoidance
- Responsible AI: fairness, transparency and governance
- Model deployment, monitoring and cloud data platforms
- Consulting-style capstone for a fictional UK energy supplier
Become a certified Generative AI Engineer.
A ~3-month professional programme covering how LLMs work, prompt engineering and structured outputs, RAG, fine-tuning, agents and tool use (including MCP), evaluation and LLMOps, safety and governance, production deployment, and an industry capstone — with hands-on JupyterLab labs in every section and a route to the VCA-GAI certification.
9 sections
- LLM foundations: tokens, embeddings, attention and transformers
- Prompt engineering and structured outputs with the OpenAI API
- RAG and embeddings: vector search, hybrid retrieval, evaluation
- Fine-tuning and adaptation, including PEFT/LoRA
- Agents, tool use and the Model Context Protocol (MCP)
- Evaluation and LLMOps: golden sets, tracing, regression testing
- Safety, security and EU AI Act-aware governance
- Production deployment: routing, caching, cost and latency budgets
- Industry capstone: a production-grade GenAI application
Both programmes share one registration flow — once you're signed up, you choose which programme's courses to start from your student portal.
Learn by doing, prove it by exam.
Learn
Work through concise, practical lessons written by domain experts — each course broken into focused modules.
Practise in JupyterLab
Every student gets a personal JupyterLab workspace in the cloud, with course notebooks and datasets ready to run in every section.
Quiz
Check your understanding after every module with instant-feedback quizzes, explanations included.
Sit the certification
Finish the curriculum, then take the Vecorra certification exam for your chosen programme — theory plus a hands-on practical lab.
The programmes support reasonable adjustments, including extra time on certification exams — see our accessibility statement.
A credential that proves you can do the work.
Every certification pairs a theory exam — 50 questions drawn from a 1,000-item bank — with a hands-on practical lab graded on real artefacts, not multiple choice alone.
Vecorra Certified Data Scientist
A 50-question theory exam drawn from a 1,000-question bank, plus a hands-on practical lab in the same JupyterLab environment used throughout the programme — a verified, employer-recognised credential.
Vecorra Certified Generative AI Engineer
A 50-question theory exam drawn from a 1,000-question bank, plus a hands-on practical lab in the same JupyterLab environment used throughout the programme — a verified, employer-recognised credential.
Your own JupyterLab workspace, in the cloud.
No setup, no installs. Open your personal workspace from any lesson and the course notebooks and datasets — including OpenAI-powered labs on the Generative AI programme — are already there, in the same environment used in the Vecorra practical exam.
Register as a student# work/training/machine-learning/03_tree_ensembles.ipynb
from sklearn.ensemble import RandomForestClassifier
from sklearn.model_selection import cross_val_score
model = RandomForestClassifier(n_estimators=300)
scores = cross_val_score(model, X, y, cv=5, scoring="roc_auc")
# mean AUC: 0.91 — nice. now try tuning max_depth…
Ready to start your AI career journey?
Join the training platform today — learn at your own pace, practise in a real environment and finish your chosen programme with a credential employers trust.