Current focus: Applied AI Masterclass
Goal: Get good at designing and shipping production AI systems (agents, RAG, evals), not just calling LLM APIs. Resource: Applied AI Masterclass by Arpit Bhayani
Working through the cohort: agentic fundamentals into tool use, memory architectures, eval frameworks, RAG, multi-agent systems, and workflow engines, through a production lens of cost, latency, guardrails, and failure modes.
What I’ve learned so far
- Evals before scale: a good eval framework is what lets you change prompts and models without fear.
- Memory is a design decision: memory architectures and write strategies shape agent behaviour more than the model choice.
- Production is the system design: cost, scalability, governance, and guardrails are the core engineering, not afterthoughts.
Next up
- Model Training: Exploring how to tre-rain models.
Recently completed
- System Design Course: Refreshed my knowledge on distributed systems and scalability patterns.