About Archestack
An independent guide to open-source AI and AI coding tools, and the engineering team that builds AI apps on the stacks it ranks.
- projects ranked
- 1,560
- lists
- 2
- categories
- 33
- refreshed a day
- 2×
The guide
We rank open-source projects from public GitHub data and review each one, with its weak spots. Nobody can pay for a place, and client work never moves a rank.
We build AI apps on the stacks we rank
Assistants, agents and private AI for startups and enterprise teams, from the first prototype to production. The research behind 1560 ranked projects goes into every pick, and the result runs in your account. On a self-hosted app's page you can bring in an engineer who knows that project, by the hour or at a fixed price. We run that service ourselves, and like every paid link here it never changes a rank.
AI products, end to end
Assistants, agents and search over your own documents, from prototype to production.
Private AI on your own infrastructure
Open models, gateways and vector databases on servers you control, next to your data.
Stack review before you commit
An independent read of the tools you plan to adopt: license, health, security and fit.
AI coding workflows
Coding agents, shared rules and review gates, so your engineers ship faster and still review.
What enterprise teams get by default
| Your security review asks | What we build in | What you can inspect |
|---|---|---|
Where does our data go?Your data stays where you say | Models run on your servers or in your cloud account. When a hosted model is used, it is one that does not train on your data, and only on the data you approve. | A data-flow diagram of every system that sees your data |
Who can use it, and what is logged?Access and audit | Sign-in through your identity provider, roles per team, and a log of every model call: who asked, what was sent, what came back, what it cost. | Your sign-in setup and the log of every model call |
How do we know it works?Measured before launch | A test set built from your real questions, scored before every release, so quality is a number your team can track. | The test set, and the score of every release |
What stops it doing harm?Guardrails | Personal data masked before it reaches a model, defenses against prompt injection, and a person approving any action that changes data. | Masking rules and a prompt-injection test report |
What will it cost to run?Costs you can predict | A monthly estimate before we build, budgets per team in the gateway, caching, and smaller models where they do the job. | A monthly estimate and a budget per team |
Can we leave later?No lock-in | Open-source parts with licenses checked, each one swappable, and everything in your repositories as code. | A license check per part, and everything as code in your repository |
Who you work with
An engineering team that ships AI systems for startups and enterprise teams. We work inside your repositories, tickets and chat, and the people on your calls are the people writing the code.
- Research-led
- The research that ranks 1560 projects picks the stack for your build: is the license safe, is the project still maintained, what breaks first.
- Remote, any region
- Based in Vietnam (UTC+7), working in English with teams in any region.
- Proof in public
- This site rebuilds itself from GitHub data twice a day, and our MCP server lets Claude and Cursor query it.
Have an AI project in mind?
Fixed scope and milestones, your data stays in your account, and you own the code.