Learn AI · Browse tracks
Advanced Key Tech
Production tooling — Ray, Triton, Kafka, OpenTelemetry, TensorRT-LLM/SGLang, MLflow LLMOps.
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
- RayDistributed Python for training, batch inference, and serving — Ray Train, Data, and Serve patterns for LLM platforms.Open
- Triton Inference ServerNVIDIA Triton — multi-framework model serving, ensembles, dynamic batching, and where it sits beside vLLM / TensorRT-LLM.Open
- Kafka for evented AIAsync ingestion, embedding pipelines, and agent side-effects — using Kafka (or similar logs) so AI work is durable and replayable.Open
- OpenTelemetry for LLMsTraces, spans, and GenAI semantic conventions — debugging agents, RAG, and cost with OpenTelemetry.Open
- TensorRT-LLM and SGLangHigh-performance inference stacks beyond vanilla vLLM — NVIDIA TensorRT-LLM and SGLang’s structured generation runtime.Open
- MLflow for LLMOpsTracking prompts, chains, and models — registries, signatures, and evaluation hooks used in Databricks GenAI paths.Open