bld-201
Intermediate
Available
A hands-on, code-first course on the three major LLM APIs side by side: messages, streaming, structured outputs, tool calling, vision, cost, caching, batch, retries, abstraction layers, cloud routes and testing. You finish by building and hardening a small multi-vendor service.
bld-210
Intermediate
Available
A practitioner's course on retrieval-augmented generation: when to use it, how to ingest, chunk, embed and index documents, how to combine BM25 and vector search with reranking, and how to measure, debug and operate the result. Every module includes working Python sketches.
bld-215
Intermediate
Waitlist
A deep dive into the two layers under every semantic search system: embedding models and vector databases. Covers how embeddings are trained and benchmarked, Matryoshka truncation and quantization, ANN index internals (HNSW, IVF-PQ, DiskANN), filtering, multi-tenancy and the operational work of re-embedding and capacity planning.
bld-230
Intermediate
Available
A practical course on application-level LLM evaluation: turning product requirements into test cases, choosing between code checks, human review and LLM-as-judge, calibrating judges against people, running evals in CI with tools such as Promptfoo and Inspect, and closing the loop with production traces and online experiments.
bld-240
Intermediate
Waitlist
Build real-time conversational agents for chat and voice. Covers conversation design, dialogue state, speech-to-speech models such as OpenAI's Realtime API and the Gemini Live API, cascaded STT-LLM-TTS pipelines, turn detection and barge-in, WebRTC and SIP telephony with LiveKit, Pipecat and Twilio, latency budgets, and testing voice agents.
bld-250
Intermediate
Waitlist
Build application features on multimodal models: image understanding and visual question answering, PDF and scanned-document extraction, chart and screenshot reading, speech-to-text and audio understanding, and video analysis. Compares frontier APIs with open-weight vision-language models and teaches cost control, evaluation and the failure modes specific to non-text inputs.
bld-260
Intermediate
Waitlist
A side-by-side, code-first comparison of the main LLM application frameworks. You build one reference application (a document-grounded assistant with tools) in LangChain 1.x with LangGraph, LlamaIndex, Pydantic AI and Microsoft Agent Framework, then examine DSPy, Haystack, CrewAI and Mastra, and finish with a framework decision record and a migration plan.
bld-270
Intermediate
Waitlist
A hands-on course on the LLM gateway pattern. You deploy LiteLLM Proxy and compare it with Envoy AI Gateway, Kong AI Gateway, Azure API Management's AI gateway policies, Cloudflare AI Gateway and Portkey, then implement virtual keys, routing and fallback, token quotas and budgets, caching, guardrail hooks, logging and audit for a multi-team organization.
bld-280
Intermediate
Waitlist
An end-to-end build of a private, on-premises AI assistant. You size GPU hardware, choose open-weight models such as gpt-oss, Qwen, Mistral and Gemma, serve them with vLLM, deploy Open WebUI or LibreChat with SSO and role-based access, connect internal documents for retrieval, and run the service with monitoring, governance and a clear support model.
bld-320
Expert
Available
An engineering course on building agents that call tools reliably, speak the Model Context Protocol, coordinate with other agents and stay safe, measurable and affordable. Every module pairs concepts with working code across the major vendor SDKs.
bld-330
Expert
Waitlist
The engineering discipline of running LLM-backed features at scale: SLOs and error budgets for probabilistic systems, multi-provider failover and capacity reservations, latency engineering from TTFT to perceived speed, cost engineering with caching, batching and routing, OpenTelemetry GenAI tracing, and safe rollout of prompt and model changes.
bld-340
Expert
Waitlist
An engineering course on guardrails as a measured, layered system. Covers policy taxonomies, input and output classifiers (OpenAI moderation, Llama Guard 4, Prompt Guard 2, gpt-oss-safeguard), NeMo Guardrails and Guardrails AI, cloud guardrail services such as Amazon Bedrock Guardrails, Azure AI Content Safety and Google Cloud Model Armor, grounding and hallucination checks, action guardrails for tools, and how to evaluate, tune and operate them.
bld-350
Expert
Waitlist
An expert course on knowledge graphs for LLM applications. You learn property graphs and RDF, query with Cypher, GQL and SPARQL, build graphs from documents with LLM extraction and entity resolution, implement Microsoft GraphRAG, LazyGraphRAG-style and LightRAG-style retrieval, text-to-Cypher and hybrid graph-vector search, and evaluate when a graph actually beats vector RAG.