Connect once, work everywhere
Glen shows up where teams already work. The plugin, MCP server, and command line tool all give full access, and setup takes less than five minutes. We hold ourselves to that standard of simplicity in everything we ship.
Careers
Help us build the memory layer for AI-powered work
Glen puts the right organizational context into every prompt, so agents work better, faster, and on fewer tokens. We're hiring a Founding Member of Technical Staff in San Francisco.
Open roles
1 open roleWhat we're building
Glen captures what a company already does - agent sessions from Claude Code, Codex, and Cursor, plus Slack, GitHub, tickets, calls, and docs - and synthesizes it into one store that holds the full picture. Nothing is done by hand. Capture runs on its own.
That context goes back where it matters. Glen finds the prior work that bears on a task and steers the model with it, so equivalent work comes back significantly faster and on fewer tokens. Skills distill from a team's work and surface automatically when someone hits a task they cover. Transcripts, live session takeovers, and shared artifacts follow you across any harness or provider.
The result is a company brain you can ask anything, from Slack, any MCP agent, or the web app. For code, the PR reviewer and code search query clones of the agents that made the change, so you get the why, not just the what.
Capture, not busywork
We believe the best knowledge systems ask nothing of the people using them. Glen collects as the work happens rather than asking anyone to document it after the fact.
Public by default
Glen reads public material only, such as public Slack channels and shared team transcripts. We are deliberate about what the system sees and how it is used.
Founding-level ownership
This is an early team. The engineer we hire will shape the product, the architecture, and how we work, not just execute against a plan.
Come build it with us
If the idea of making an entire company queryable sounds like the problem you want to spend your time on, we'd like to hear from you.