Full Course Curriculum

Autonomous AI Systems &
Agent-Based Computing

12 sessions · 4 lab tests · One running project that evolves from a resume to a fully autonomous AI system.

📄 Download Session Plan (PDF) 📋 Download Full Syllabus (.docx)
Unit 1 · Sessions 1–3 Foundations of Modern AI
01
Introduction to AI, Generative AI & Agentic AI AI history, evolution ladder, three types of AI, LLM overview
Intro · Concepts · History
View Session →
02
LLM Fundamentals & Output Behavior Tokens, temperature, context windows, system messages, prompt injection
LLMs · Tokenisation · Params
View Session →
03
Prompt Engineering Fundamentals Role prompting, few-shot, chain-of-thought, production prompt design
Prompting · Few-shot · CoT
View Session →
Unit 2 · Sessions 4–5 Structured Data & Retrieval
04
Structured Outputs & Parsing JSON schema design, output formatting, error handling, pipeline construction
JSON · Parsing · Pipelines
View Session →
05
Retrieval-Augmented Generation (RAG) Embeddings, semantic search, vector databases, RAG architecture
RAG · Embeddings · VectorDB
View Session →
Unit 3 · Sessions 6–7 Specialization & Measurement
06
Fine-Tuning & Specialization Pre-training vs fine-tuning, LoRA, QLoRA, instruction optimisation, training data
LoRA · Fine-tuning · Data
View Session →
07
Evaluation & Benchmarking Metrics, benchmark creation, A/B testing, performance monitoring, failure analysis
BLEU · A/B Testing · Metrics
View Session →
Unit 4 · Sessions 8–9 Production & Optimization
08
Scaling & Production Deployment Infrastructure, horizontal scaling, caching, cost optimisation, observability
Infra · Scaling · Cost
View Session →
09
Advanced Techniques & Optimization Ensembling, prompt optimisation, knowledge distillation, retrieval optimisation
Ensembling · Distillation · DSPy
View Session →
Unit 5 · Sessions 10–12 Safety, Agentic AI & Capstone
10
AI Safety & Alignment Bias detection, explainability, human-in-the-loop, AI governance & regulation
Safety · Bias · Ethics
View Session →
11
Advanced Architectures Multi-agent systems, tool use, memory, planning, agent frameworks
Multi-Agent · ReAct · Tools
View Session →
12
Real-World Case Studies & Capstone Production architectures, recommendation systems, autonomous systems, capstone project
Case Studies · Capstone
View Session →
Course Objectives

What Students Will Be Able To Do

✓Understand the fundamentals of Generative AI and how Large Language Models work at a mechanistic level
✓Design effective prompts for diverse production use cases using systematic engineering techniques
✓Build AI-powered applications using both proprietary APIs (OpenAI, Anthropic) and open-source tools
✓Architect Retrieval-Augmented Generation systems with embeddings, vector databases, and semantic search
✓Evaluate AI systems rigorously — create benchmarks, run A/B tests, and monitor production performance
✓Deploy AI systems to production with appropriate infrastructure, scaling, and cost optimisation strategies
✓Assess AI systems for bias, explain decisions, and implement responsible AI governance practices
✓Design multi-agent systems that plan, use tools, maintain memory, and act with genuine autonomy
Begin the Journey

Start With Session 1

All sessions, presentations, and study notes are ready. Start whenever you are.

Begin Session 1 → View Lab Tests