dev-201
Intermediate
Available
A code-first course on integrating LLMs into Python the way production code needs it: uv projects and secrets, the OpenAI, Anthropic and Google GenAI SDKs side by side, async concurrency and streaming, Pydantic structured outputs and tools, retries, local models, embeddings and RAG, FastAPI and pandas integration, testing and evals with pytest, observability, cost tracking and deployment.
dev-202
Intermediate
Waitlist
Call OpenAI, Anthropic, Gemini and local Ollama models from Perl with real CPAN modules, add streaming, concurrency and validated JSON output to text pipelines, and use AI coding assistants to modernize legacy Perl behind characterization tests.
dev-203
Intermediate
Waitlist
Build production AI features in TypeScript: the official OpenAI, Anthropic and Google Gen AI SDKs, the Vercel AI SDK for provider-neutral streaming and tools, Zod-typed structured outputs, in-process models with Transformers.js and node-llama-cpp, and tests that never hit a live API.
dev-204
Intermediate
Waitlist
Integrate LLMs into JVM services with Spring AI 2.0 and LangChain4j, the official OpenAI, Anthropic and Google Gen AI Java SDKs, and Kotlin-native Koog: typed structured output, tool calling and MCP, retrieval, streaming with Reactor and coroutines, virtual threads, local models, and tests with Testcontainers and WireMock.
dev-205
Intermediate
Waitlist
Build AI services in Go with the official openai-go, anthropic-sdk-go and google.golang.org/genai SDKs: streaming over SSE, structured output with generated JSON Schema, tools and the official MCP Go SDK, errgroup and rate-limited concurrency, local inference through Ollama, yzma or ONNX Runtime, and testing and container deployment.
dev-206
Intermediate
Waitlist
Integrate LLMs into .NET 10 applications: the IChatClient abstraction in Microsoft.Extensions.AI, the official OpenAI, Anthropic and Google.GenAI SDKs, typed structured output, IAsyncEnumerable streaming, Microsoft Agent Framework and MCP, local inference with ONNX Runtime GenAI, Foundry Local and Ollama, plus testing, resilience and deployment.
dev-207
Intermediate
Waitlist
Work with AI where the data already lives: vector types and indexes with pgvector and SQL Server 2025, hybrid search, in-database model calls in AlloyDB, Snowflake, BigQuery and Databricks, embedding pipelines that stay in sync, and text-to-SQL with schema grounding, read-only guardrails and evaluation.
dev-208
Intermediate
Waitlist
Use LLMs from R with ellmer and httr2: structured extraction into data frames, parallel and batch calls, local models through Ollama, retrieval with ragnar, evaluation with vitals, Shiny chat apps, and reproducible, tested workflows with renv, testthat and recorded HTTP.
dev-209
Intermediate
Waitlist
Scientific machine learning in Julia: universal differential equations, neural ODEs and physics-informed networks with SciML, Lux and ModelingToolkit; automatic differentiation with Enzyme, Zygote and Mooncake; GPU and multithreaded training; surrogates and symbolic regression; LLM calls with PromptingTools.jl; and testing and packaging research code.
dev-210
Intermediate
Waitlist
Script LLM calls from Bash and PowerShell with proper JSON handling, retries and secrets hygiene; drive terminal coding agents in headless mode; and add review, triage and generation steps to CI/CD pipelines with guardrails that keep generated commands from running unchecked.
dev-211
Intermediate
Waitlist
Build on-device AI features on both platforms: Apple's Foundation Models framework and Core ML on iOS, ML Kit GenAI with Gemini Nano and LiteRT on Android, and cross-platform runtimes such as ExecuTorch and ONNX Runtime. Covers model conversion, quantization, device-capability checks, hybrid cloud fallback, performance and privacy.
dev-212
Intermediate
Waitlist
Use coding assistants and MCP-connected agents to write, refactor and review Terraform/OpenTofu and Ansible, then prove the result with plans, tests, linters and policy as code. Covers why IaC is unusually risky for hallucinated arguments and stale provider knowledge, and how to keep agents read-only until a human approves.
dev-213
Intermediate
Waitlist
Build and govern AI inside the three big enterprise platforms: Agentforce agents, prompt templates and Apex on Salesforce; Now Assist skills and AI agents on ServiceNow; Joule agents and generative AI hub apps on SAP BTP. Then connect them through MCP, A2A and APIs without losing data permissions or audit.
dev-301
Expert
Available
Expert-level Rust for AI: in-process inference with Candle, Burn and ONNX Runtime (ort), bindings to llama.cpp through FFI, GPU integration with CUDA, PyO3 bridges to Python, and high-performance async services with batching, backpressure and observability. Emphasizes unsafe-code discipline, zero-copy data paths and measured performance.
dev-302
Expert
Waitlist
Embed LLM and classic model inference directly in C programs: the llama.cpp/ggml C API and GGUF models, the ONNX Runtime C API, CUDA from C, memory and threading discipline, and the testing and fuzzing needed to make native inference code safe to ship.
dev-303
Expert
Available
Expert C++ for AI inference: exporting PyTorch models for C++ with torch.export and AOTInductor, LibTorch, the ONNX Runtime C++ API, building and running TensorRT engines, writing CUDA kernels and custom plugins, and designing concurrent, memory-safe inference services with measured performance.
dev-304
Expert
Waitlist
Couple trained neural-network surrogates to Fortran and C++ HPC simulations: choosing what to replace, generating training data from the simulation, coupling in-process with FTorch, TorchFort, Fiats, LibTorch or ONNX Runtime via iso_c_binding, running inference on GPUs alongside MPI ranks, online training, and validating stability, conservation and scaling.
dev-305
Expert
Waitlist
Design event-driven architectures that call LLMs and ML models: Kafka topics, consumer groups and share groups, queues and DLQs, Flink SQL model inference, CDC-driven enrichment and embedding pipelines, durable agent workflows, and the idempotency, rate-limit, ordering and cost controls that make AI calls safe inside streams.