Launching today
Parallax
Every coding agent you run, in one place.
76 followers
Every coding agent you run, in one place.
76 followers
Parallax runs your coding agents for you. Claude Code, Codex and 40+ more in one app, with 240+ models on your own subscriptions. Captain handles the busywork, fast merge ships your PRs in one click, and when a subscription runs out, auto fallback enables your work to move to the next.
Super excited to be a part of this project and share Parallax with everyone.
I usually have a ton of agent sessions running at once, and for the longest time I was the one keeping track of all of them. I was also constantly juggling subscriptions, and so whenever one would hit its limit mid-task, I would have to stop, switch accounts and pick it back up. On top of that, every time a new model dropped I’d move my whole workflow (Codex when GPT-6 Astra came out, then back to Claude when Opus 5.5 came out).
The separate apps, separate sessions, separate everything were a huge hassle. Parallax fixed all of that for me. When one subscription runs out, it moves the work to the next one and my agents just keep going. And every model is in one place, so switching to whatever just came out takes a second. But the thing I’m most excited to share is Captain. It keeps track of all my sessions for me and makes sure each one stays aligned with what I’m actually trying to do. If an agent starts drifting, Captain steers it back so it answers the prompt and hits the objective. I finally don’t have to watch every session myself.
We just launched, and I’d love for you to try it
Captain steering a drifting agent back to the objective is the part that caught my eye! Watching every session yourself is what makes running several at once so tiring. How does Captain decide that an agent has drifted, and when it steers one back, can you see what it told the agent?
@neerajnathany Captain checks what each agent is doing against the task you gave it. It looks for signs that an agent is stuck, going in circles, duplicating work, failing tests, or getting off track. And yes, you can see exactly what it said! Every steer includes the reason behind it and the message sent to the agent.
The subscription fallback is the detail I like here. I'd want the handover to stay visible so I can tell which agent picked up the work and why.
@oleksii_sekundant Usually, the same agent keeps working. If one subscription runs out, the session switches to your next one with the conversation intact. You’ll see a note in the session, like “Moved to Claude 3 · Claude signed out.” A different agent only takes over after a set number of attempts to fix failing checks. When that happens, you’ll see something like “Codex fixing it.”
Congrats Charles and the Conifer team on the Parallax launch. Having reviewers always come from a different lab than the one that wrote the code is a useful constraint most tools skip. When the sandbox later refutes a claim the agent made in the session transcript, does that feed back into Captain as session drift on the live agent, or only show up as a labelled finding on the review PR?
@genedai It goes straight to the agent that’s working on the task. If Parallax’s own test run fails after a turn, it adds a clearly labeled message to the session explaining what failed. Captain recognizes that as a test failure rather than the agent going off track. Review findings are labeled on the PR, and any unresolved review threads get sent back to the responsible agent.