Orchestra is the shared workspace for your team and its coding agents. Run as many agents as you need, watch any session live, step in or fork it, and hand work to a person or an agent. Your newest engineer starts where your best one finished.
The app, with demo data
Works with the agents you already runClaude CodeCodexGemini CLICursorOpenCode
Recorded on the real product: a team and its agents fixing one bug together, with what the team knows in every session.
Code used to be the shared object. Now it is people, then agents, then code, and the middle step happens in a terminal nobody else can open.
Arnav’s terminal
why does the ingest test flake?
the clock is frozen in setup
Only you can see this
Eashan’s terminal
retry the webhook handler
tried it, the race is still there
Only you can see this
Maya’s terminal
where do we redact keys?
looking at session context
Only you can see this
Arnav’s terminal
why did staging run out of VMs?
nothing caps how many one team can start
Only you can see this
An illustration: four agent sessions, each in a terminal only one person can see
Orchestra puts every agent session where your team can see it, and hands what it learned to the next one.
The agent session is where the work happens now. Orchestra opens it to your team.
1Your Agents Work
Sessions come from the agents you already run. Secrets are stripped on your machine.
2Your Team Joins
Watch any session live, step in mid-task, fork it, or hand the work to a person or an agent.
3Orchestra Keeps the Record
Every decision, dead end and open question, linked to the moment it happened.
4The Next Session Starts Smarter
From earlier sessions
Decision: inject the clock
Dead end: retries hide the race
Whoever starts next, person or agent, starts with what the team already knows.
Every session leaves the team knowing more
Run as many agents as you need, side by side, across every engineer on the team. Every running session is in one list: whose it is, what it is on and how long it has been going.
The app, with demo data: every engineer's running agents in one list
Open any running session in your workspace from your own machine, follow along, and step in mid-task, or fork it to try another way. Everyone sees who said what.
The app, with demo data: Eashan’s message in a session Claude Code is running for Arnav
People and agents share one assignee list. Give an issue to an agent and it starts with what the team already knows: earlier attempts, dead ends and decisions.
Claude Code starts with
The app, with demo data, and what the agent starts with
Every session ends with a written record of what it learned: what worked, what is still open and what failed. It stays on the issue, linked to the moment it happened.
The app, with demo data: a finished session’s conclusion
Orchestra hands each new session what earlier ones learned, tried and ruled out. Your agents stop starting from zero, and your newest engineer works with what your best one learned.
EVEashan’s sessionDone
Decision
Inject the clock at the policy boundary. Never freeze time in a test.#4111
ARArnav’s sessionDone
Dead end
Retries hide the race.#4116
Claude CodeMaya’s terminal
> the export test flakes on CI, fix it
Context from previous work in this repo
Same flake as W-214. Injecting the clock instead of adding a retry.
Whoever starts the next session, person or agent, starts with what the team knows
Matched to the repo, the branch and the issue, so the agent is not buried in history
Each item points at the moment it was learned, so nothing is taken on faith
Keep the agents and tools you use today. Orchestra connects them.
Conversations exported from claude.ai and ChatGPT import the same way.
Sessions, issues, commits and decisions link to each other automatically. Open one and the rest come with it. Nobody has to tag anything.
Orchestra keeps a record of the work that is never edited: who did what, and for whom. Everything it learned links back to it, so you can check instead of trust.
The app, with demo data
A person, or an agent and the person it worked for
Nothing in the record is rewritten after the fact, so what you check is what happened
Specs and docs join the record, linked like everything else
Across the sessions we run on Orchestra ourselves, the median session leaves five decisions and two approaches that failed.
Before: Agent sessions live in one person’s terminal
With Orchestra: Every session is open to your team, live
Before: Getting help means re-explaining in Slack
With Orchestra: A teammate opens your session and steps in
Before: A handoff is a branch and a summary from memory
With Orchestra: Give the issue to a person or an agent, and they start with everything
Before: What was tried leaves when the terminal closes
With Orchestra: Decisions, open questions and dead ends stay on the issue
Before: Every agent starts from zero
With Orchestra: Every session starts with what your team already knows
Before: Newer engineers learn by asking around
With Orchestra: They start where your best engineer finished
Orchestra is in private beta. Leave your work email and we’ll set up your workspace, connect the agents your team already runs, and get you into your first shared session.