For Go engineers who need to build fast, reliable services around LLMs: official SDKs, streaming, structured output, tools and MCP, bounded concurrency, local inference, and small, secure deployments.
The curriculum is being prepared and reviewed. Enrollment is not open yet.
About this course
Go is a natural fit for the service layer around language models: gateways, proxies, agent backends, MCP servers and high-throughput batch workers. Its concurrency model, small static binaries and strong standard library suit workloads that spend most of their time waiting on network calls and streaming bytes.
This course teaches Go AI integration with current libraries. You will use the official SDKs from OpenAI, Anthropic and Google, stream tokens through net/http with server-sent events and context cancellation, and generate JSON Schema from Go structs for structured output with validation. You will build tool-calling loops and MCP servers and clients with the official Model Context Protocol Go SDK, and compare frameworks (Genkit Go, Google's Agent Development Kit for Go, Eino and LangChainGo) against plain SDK code. You will manage concurrency with errgroup, semaphores and golang.org/x/time/rate, and build an AI gateway with retries, quotas and audit logging. For local inference you will use Ollama's Go API, in-process llama.cpp through yzma, and ONNX Runtime bindings. Finally you will test with httptest, recorded fixtures and testing/synctest, and ship minimal containers with ko or distroless images. Versions are as of October 2026 (Go 1.27).
What you will be able to do
Call OpenAI, Anthropic and Gemini from Go with the official SDKs, with context deadlines, retries and error classification
Stream model output through an HTTP handler with server-sent events, flushing and client cancellation handled correctly
Generate JSON Schema from Go structs for structured output and validate responses before use
Build an MCP server or client with the official Go SDK and implement a tool-calling loop
Implement bounded, rate-limited concurrent processing with errgroup and x/time/rate
Run a model locally from Go through Ollama or in-process inference and justify the trade-offs
Test AI-dependent Go code deterministically and ship it as a small, observable container
Syllabus
6 modules · 17 lessons · about 12 hours
1
Module 1: Official SDKs and Idiomatic Calls
Calling the major providers the Go way.3 lessons
openai-go, anthropic-sdk-go and google.golang.org/genai: clients, options and errors
Context, deadlines, retries and error classification
Configuration, secrets and cost accounting per request
2
Module 2: Streaming and Structured Output
Moving tokens and typed data safely.3 lessons
Streaming responses through net/http with server-sent events and flushing
JSON Schema from Go structs with invopop/jsonschema, and validating model output
Handling partial, truncated and malformed responses
3
Module 3: Tools, Agents and MCP
From function calling to full agent backends.3 lessons
Tool-calling loops with plain SDK code
MCP servers and clients with the official Model Context Protocol Go SDK
Frameworks compared: Genkit Go, ADK for Go, Eino and LangChainGo
4
Module 4: Concurrency and AI Gateways
High-throughput services that respect provider limits.3 lessons
Fan-out with errgroup, semaphores and bounded worker pools
Rate limiting with x/time/rate, backoff and quota management
Building an AI gateway: routing, caching, keys and audit logs
5
Module 5: Local Inference from Go
Models on your own hardware, over HTTP or in-process.2 lessons
The Ollama Go API and OpenAI-compatible local servers
In-process inference with yzma and ONNX Runtime bindings
6
Module 6: Testing and Deployment
Shipping AI services in Go with confidence.3 lessons
Testing with httptest, recorded fixtures and testing/synctest
Integration tests against local models with testcontainers-go
Small, secure containers with ko and distroless, plus OpenTelemetry tracing
Exam and certificate
The exam unlocks once you have completed every lesson. It draws 20 questions from a larger bank, spread across the modules, with the answer options shuffled for each attempt. You have 40 minutes and need 80% to pass. After a failed attempt you can try again after 24 hours, up to 3 attempts in total. Feedback is given per module, never as the answer key.
Passing earns a certificate of completion and assessment from the Wirewalk Institute of AI with a unique ID that anyone can check at institute.wirewalk.com/verify. It is not an accredited qualification or a vendor certification. How certification works.
The Wirewalk Institute of AI is independent. This course teaches publicly documented products and is not affiliated with or endorsed by OpenAI, Anthropic, Google (Gemini, Genkit, Agent Development Kit), Ollama, CloudWeGo (Eino), LangChainGo, ONNX Runtime, or Model Context Protocol. Product and company names are trademarks of their respective owners.
A free, one-hour, hands-on start built on one realistic job: sorting a property manager's 1,150 monthly maintenance requests. You get an API key the safe way, make real calls with curl and an official Python SDK, run them over forty work orders, read stop reasons, usage and errors, estimate the monthly cost, and decide whether the script is ready. Exercises run on your own provider account for well under a dollar.
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.
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.