// learning in public
Working notes from going deep on AI engineering (LLM orchestration, agents, RAG, and the foundations under them) while sharpening the fundamentals: algorithms and system design. Patterns, trade-offs, runnable code, and progress tracked in the open.
Engineering with large language models: orchestration, agents, RAG, and LLMOps.
15 topics · 4 notes
Foundations underneath the tools: representations, training objectives, and the math of meaning.
1 topic · 1 note
Data structures and algorithms for interviews and beyond: patterns, proofs, and practice problems.
12 topics · 3 notes
Designing large systems: building blocks, trade-offs, and worked case studies.
8 topics · 0 notes
A model can't do anything; it can only emit JSON that asks for something to be done. How tool schemas reach the model, how the application executes the calls and feeds results back, and what LangGraph's ToolNode handles for you.
A checkpointer remembers one thread. Agents also need to remember users across sessions and search corpora far larger than any context window. Where each kind of memory lives, and how to wire both into a LangGraph agent.
How an embedding model turns text into a vector, how it learns similarity, how vectors are searched at scale, and how semantic chunking decides what gets a vector.
Linear chains are DAGs. Agents need cycles, shared state, and checkpoints. A look at the abstraction shift.
What retrieval-augmented generation is, the indexing and answering pipelines step by step, a worked example with real similarity scores, and where RAG breaks.