Start Here

Reinforcement Learning for LLMs

From PPO fundamentals to GRPO and GDPO — the complete policy optimization series.

1. Part 1: RL Foundations
2. Part 2: PPO Deep Dive
3. Part 3: GRPO
4. Part 4: GDPO

View series landing page →

Contextual Bandits

A 5-part series from decision framework to production deployment.

1. Part 1: The Decision Framework
2. Part 2: Theory & Regret Bounds
3. Part 3: Algorithm Guide
4. Part 4: Neural Bandits
5. Part 5: Production Deployment

View series landing page →

Recommendation Systems

From contextual bandits for personalization to retrieval-augmented generation.

• Contextual Bandits in Production
• RAG System with LlamaIndex, Elasticsearch & Llama3

LLM Evaluation

Moving beyond vibes to systematic evaluation.

• Beyond the Vibe Check