LangChain
LangChain is a framework for building applications on top of language models. It supplies the plumbing: document loaders, text splitters, retrieval wrappers, tool and agent abstractions, and integrations with most model providers and vector stores, including Chroma from this library. For teams that would otherwise hand-roll the same glue code, it offers ready-made building blocks with a very large ecosystem behind them.
View the repo · langchain-ai/langchain ↗Developers prototyping AI features who want assembled components rather than a blank file, and teams whose problem genuinely spans several models, stores and tools. It rewards people comfortable reading source code, because the abstractions run deep. For a single model call with a well-crafted prompt, the provider SDK alone is simpler; reach for a framework when orchestration is genuinely the hard part of your workflow.
Install itpip install -U langchain langchain-communityThe framework moves quickly and interfaces change between releases, so pin your versions and read the changelog before upgrading. It is easy to stack abstractions until nobody can say what prompt actually reached the model, which makes debugging painful; keep chains shallow and log the raw inputs and outputs. Start with the smallest set of pieces that solves the workflow, and resist adding agents where a fixed sequence of steps will do the job.
Where Blash AI comes inWe use frameworks sparingly and deliberately. Where LangChain fits, we keep the design shallow, add the controls a business needs around cost, logging and failure handling, and hand over code your own developers can read and maintain. Where it does not fit, we say so and build the thinner version, often as plain model calls routed through LiteLLM with n8n handling the orchestration.
When you want this running on your real stack, that is the engagement
Book an AI audit →n8n →
n8n is workflow automation you run on your own infrastructure. It connects the applications a business already uses through a visual editor, with hundreds of prebuilt nodes, a code step for anything bespoke, and native support for calling language models mid-flow. Think of it as a self-hosted alternative to Zapier or Make that you own outright, including every piece of data passing through it.
Ollama →
Ollama runs open-weight language models on your own machine or server. One command downloads a model and serves it behind a local API that most existing AI tooling can talk to, so an application built against a hosted provider can point at your own hardware instead. Sensitive data stays inside your network, there are no per-token fees, and it keeps working with no internet connection at all.
LiteLLM →
LiteLLM is a gateway that lets you call models from Anthropic, OpenAI, Google and many other providers through one consistent API. It runs as a Python library inside your code or as a standalone proxy server in front of your whole stack, with cost tracking, per-key budgets, rate limits and automatic fallbacks for when a provider has a bad day.