Your Own AI Assistant — A New Service from Anlek
We use an AI assistant to run our business every day. Now we're offering to set one up for you.
AI agents your team can trust with real work.
Anlek helps small teams put AI assistants and agents to work safely. That means an AI strategy that fits how you operate, agents that act for you without holding your passwords or API keys, and shared rules that keep their work reviewable. Behind that is 25 years of production code in Rails and Elixir.
Your agent
Drafts an email or opens a pull request
Credential gateway
Holds the keys; the agent never sees them. Each call is allowed, blocked, or held for your approval, and logged.
Your accounts
GitHub, email, calendar, your own apps
AI strategy and agent setup
Which jobs in your team are worth handing to an agent, and the guardrails and shared rules that keep the result reviewable.
How we set up a teamYour own AI assistant
A personal assistant on your own hardware, reached through the chat apps you already use. We install it, connect it to your apps, and show you how to use it.
How the setup worksRails and Elixir engineering
Rails upgrades, Phoenix and Rails applications, and legacy rescue. We have done this work for 25 years.
Rails and Elixir servicesYour team stays in control
An agent that can act for you needs limits you set. In the setup we recommend, the agent never holds a credential; sensitive actions such as outbound mail or a deletion wait until a person approves them; every call is logged; the team's rules are written down once and every agent follows them; and an independent review happens before an agent's work counts as done.
We run our own work this way, with agents built on open-source tools such as OpenClaw, Hermes and Paperclip, and we try new ones as they appear.