Learn AI · Browse tracks
Key Tech
Everyday tools — LangGraph/LangChain, Hugging Face, Ollama, MCP, LlamaIndex, OpenAI/Anthropic APIs, Chroma, Weights & Biases.
How this relates Browse here for Hello Interview–style articles. The guided path below in the sidebar remains the project milestone sequence.
Articles in this track
- LangGraph and LangChain patternsGraphs, state, checkpoints, interrupts, and when frameworks help — after you can hand-roll a ReAct loop. Architecture, patterns, ops, evals, and production failure modes end-to-end.Open
- Hugging FaceHub models, Transformers, datasets, and Spaces — the open-weight ecosystem AI engineers live in. From Hub ID to chat template, PEFT, licenses, serving handoff, CI pins, and production failure modes.Open
- OllamaLocal model runner for laptops — Modelfiles, OpenAI-compatible API, quantization, CI, ChatClient swaps, and honest limits vs GPU servers — end-to-end interview depth.Open
- Model Context Protocol (MCP)A standard way for hosts (IDEs, agents) to discover and call tools/resources from MCP servers — architecture, transports, authZ, schemas, evals, and production blast radius end-to-end.Open
- LlamaIndexData framework for LLM apps — indexes, retrievers, query engines end-to-end; ingest/upsert, quality loops, synthesizers, ACL filters, debugging, and when to use vs LangChain/LangGraph.Open
- OpenAI and Anthropic APIsChat Completions / Messages APIs, tools, structured outputs, streaming end-to-end — client contract, gateway patterns, retries, routing, token accounting, and failure modes.Open
- ChromaDeveloper-friendly vector database for local RAG — collections, metadata filters, upsert discipline, metrics, ops, debugging, and when to graduate end-to-end.Open
- Weights & BiasesExperiment tracking for prompts, fine-tunes, and evals end-to-end — runs, tables, artifacts, lineage, promote gates, and how LLM teams compare variants.Open