Learning in public: LLM engineering, AI foundations, algorithms, and system design, with runnable code and progress tracked in the open.https://prabesh.net/https://prabesh.net/llms/tool-calling/execution-loops/https://prabesh.net/llms/tool-calling/execution-loops/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.Thu, 29 Oct 2026 00:00:00 GMTtool-callingagentslanggraphpydantichttps://prabesh.net/llms/core-abstractions/state-vs-memory/https://prabesh.net/llms/core-abstractions/state-vs-memory/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.Mon, 26 Oct 2026 00:00:00 GMTlanggraphmemoryraghttps://prabesh.net/ai/representations/embeddings-and-chunking-under-the-hood/https://prabesh.net/ai/representations/embeddings-and-chunking-under-the-hood/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.Thu, 22 Oct 2026 00:00:00 GMTragembeddingschunkinghttps://prabesh.net/llms/core-abstractions/chains-vs-state-graphs/https://prabesh.net/llms/core-abstractions/chains-vs-state-graphs/Linear chains are DAGs. Agents need cycles, shared state, and checkpoints. A look at the abstraction shift.Mon, 19 Oct 2026 00:00:00 GMTlanggraphorchestrationhttps://prabesh.net/llms/rag-fundamentals/rag-how-it-works/https://prabesh.net/llms/rag-fundamentals/rag-how-it-works/What retrieval-augmented generation is, the indexing and answering pipelines step by step, a worked example with real similarity scores, and where RAG breaks.Thu, 15 Oct 2026 00:00:00 GMTragretrievalembeddingshttps://prabesh.net/algorithms/two-pointers/two-pointers/https://prabesh.net/algorithms/two-pointers/two-pointers/Two pointers on sorted arrays and on linked lists: pair sums, 3Sum, middles, cycles, and intersections, each from brute force to optimal with time and space complexity.Fri, 08 Aug 2025 00:00:00 GMTtwo-pointerslinked-listarrayshttps://prabesh.net/algorithms/big-o/big-o-and-recurrences/https://prabesh.net/algorithms/big-o/big-o-and-recurrences/Asymptotic notation, how to solve divide-and-conquer recurrences with the Master Theorem, and Pow(x, n), Sqrt(x), and merge sort, each from brute force to optimal with time and space complexity.Thu, 24 Jul 2025 00:00:00 GMTbig-orecurrencesbinary-searchsortinghttps://prabesh.net/algorithms/arrays-strings/arrays-and-strings/https://prabesh.net/algorithms/arrays-strings/arrays-and-strings/Hash maps, prefix products, in-place writes, reversal tricks, and sliding windows: twelve array and string problems, each solved three ways with time and space complexity.Mon, 21 Jul 2025 00:00:00 GMTarraysstringshashingsliding-window