LangChainGo: Building AI Applications with Golang

Discover LangChain's unified API for building LLM apps across models like OpenAI GPT and Gemini, leveraging prompt templates, vector databases, and AI agent tools without rewriting code.

Explore LangChain components, including chat models, prompt templates, message types, and output parsers. Review document loaders, text splitters, embeddings, vector stores, retrievers, tools, memory, AI agents, retrieval augmented generation.

Develop a basic understanding of Golang, secure access to an LLM (OpenAI or Gemini), and prepare a code editor to start the LangChain Go course.

Learn how to obtain and securely manage API keys for OpenAI and Google Gemini, including creating keys, naming, permissions, and safeguarding credentials with environment variables.

Build your first LangChainGo application by initializing a Go module, installing the long chain go and OpenAI wrappers, and running a main file that prompts OpenAI and prints the response.

Master string prompt templates in LangChainGo, learn how a template string and input variables produce output using a value map, Go's text templating, and ginger templates and f strings.

Master standard template definition for prompt templates in Golang, using a prompt template struct with template, input variables, and a Go template format to drive dynamic topic and website inputs.

Learn to apply Ginger templates in LangChain Go prompt templates, following Jinja syntax with dynamic variables, and set the template format to Jinja2 to translate statements from language to another.

Create multi-line templates in Golang using backticks, with if and range logic and few-shot examples to format a list of professionals and their clubs.

learn to use partial variables to prefill a prompt template in go, enabling dynamic date and company name before sending to the language model.

Learn to format a prompt template in Go, pass it to an LM with the OpenAI GPT-4 model, and retrieve a live response via the LMS pipeline.

Learn how LangChain Go structures prompts with system, user, and assistant messages to guide AI behavior across a chat session. Explore chat prompt templates and example conversations to shape responses.

Build a multi-message chat prompt template by creating system and human message templates, combine with a prompt formatter, and test translating to French using an OpenAI LLM.

Explore how chat models differ from traditional LLMs, enabling conversational memory, structured outputs like JSON, multimodal inputs and real-world actions, with a hands-on Go demo using long chain go.

Learn to instantiate Gemini models in Golang using LangChainGo, including setting the Google API key and switching from OpenAI to Google AI packages.

Configure models in LangChainGo by setting API key, model, response format, and base URL. Learn to manage environment variables, local models, and provider specifics for OpenAI and Gemini.

Configure model responses by adjusting temperature and max tokens, and use options like with model, top K, top P, stopwords, functions, and tools in LangChain Go's LMS generate calls.

Discover how to run local AI models on your computer with Ollama, weighing benefits and hardware trade-offs, and integrate local models into LangChain Go workflows without API keys.

learn to use a fake LLM for local development, prefill with dummy responses to avoid token costs and latency, and swap in a real LLM for testing.

Discover how chains link predefined steps to build AI workflows, passing outputs as inputs between steps. Learn lm chain, sequential chain, and conversational chain with memory and their use cases.

Build your first LM chain by linking a prompt template to an OpenAI language model to translate text between languages, using GPT four and the chains package.

Build a workflow with a sequential chain linking a synopsis chain, a character chain, and a backstory chain to turn a topic into a story and backstory using LM chains.

Build a command-line chat app that uses a conversational chain with a memory module and in-memory buffer to remember conversation history, powered by LangChain Go and GPT-4.

Explore how agents differ from chains, using a large language model for thinking, tools for actions, and memory, prompt templates, and an executor to orchestrate execution.

Explore how the react pattern powers AI agents in long chain go by modeling thought, action, and observation loops, using tools to iteratively complete tasks.

Create your first react agent in LangChain Go using the one-shot agent and a calculator tool. Configure GPT-4, a tool set, and a five-iteration executor to solve math problems.

learn to build a custom word counter tool in Go using the long chain tool interface, and run it with an AI agent alongside a calculator tool.

Build a web-browsing AI agent with LangChain Go and the SAP API to extend knowledge beyond model cutoffs, using a simple context and web search workflow.