Training

AI Engineer

Build AI applications that retrieve information, use tools, and return answers with sources. The track covers language models, retrieval, agents, deployment, and evaluation.

Before you start

Python, APIs, Git, SQL, JSON, and REST fundamentals.
Basic machine learning concepts and model evaluation.
Comfort reading framework documentation and working with unfamiliar libraries.
Time to build and document several applications.

Roadmap / 18 weeks

Weeks 1–4 / Machine learning and LLM foundations

Project: Prompt-evaluated assistant

Weeks 5–8 / Embeddings, vector databases, and retrieval

Project: Document-grounded chatbot

Weeks 9–12 / LangChain, LangGraph, and agents

Project: Stateful tool-using agent

Weeks 13–18 / Deployment, evaluation, and safety

Project: Deployed AI application

Syllabus

Machine learning foundations

Supervised and unsupervised learning, deep learning concepts, model evaluation, Python APIs, and reproducible experiments.

Project: Compare a classical baseline and a deep learning model on the same classification task.

Language models and prompt engineering

Tokens, context windows, structured outputs, prompt patterns, test sets, and failure analysis.

Project: Build an assistant for a business workflow and evaluate its prompts.

Embeddings and vector search

Chunking, metadata, semantic and hybrid search, reranking, and vector databases.

Project: Build and evaluate semantic search over a collection of documents.

Retrieval-augmented generation

Document ingestion, retrieval, answers with source citations, and evaluation of quality, latency, and cost.

Project: Build a document-grounded chatbot and write an evaluation report.

LangChain, LangGraph, and agents

Tool calling, workflow state, routing, checkpoints, retries, and human approval.

Project: Build a multi-step agent with tool use, logs, and safe fallback behavior.

Backends and deployment

FastAPI or Node, streaming, caching, authentication, rate limits, logging, and feedback.

Project: Deploy a RAG or agent application with monitoring and feedback collection.

AI safety and governance

Prompt injection, data handling, access controls, evaluation gates, model cards, and audit logs.

Project: Add security controls, a threat model, and adversarial test cases.

Schedule and fees

Enquire about the next start date, weekly schedule and total fees before enrolling.

Send an enquiry →