Anthropic's developer account posted this on 17 September, and the problem it is aimed at is one anybody running several Claude Code sessions at once already knows: you decide what each one works on, repeat the same background at the start of each, and check back to see which finished or needs an answer. In a project that coordinating moves to Claude. One long-running conversation takes what you send, a bug report, a stack trace, a list of tasks, and starts a thread for each piece of work or hands it to the thread already working in that area. Each thread is a cloud session with its own context window, on its own branch, opening a pull request when the work calls for one and reporting back when it finishes, and every new thread starts with the project's repositories, instructions and memory, so a rule stated once, such as which branch to target, reaches all of them. An Overview pane lists which threads finished, which pull requests are ready for review and which thread is waiting on your answer, which is the concrete form of the line about it keeping going when you leave. The examples in Anthropic's own docs are a lint config brought across every service, one service's stream of bugs and review requests pasted in as they arrive, a migration bigger than one session, and a folder of contracts or a support-ticket export you keep asking new questions of. The limits are stated just as plainly. Threads work on GitHub repositories and uploaded files, not on anything only your machine can reach, so a local database or a device emulator still means a local session; it is on Pro and Max and not yet on Team or Enterprise; a project uses the same plan limits and uses them faster, with a ceiling of 200 new threads a day; and the name is a redesign of the Projects that already existed in the claude.ai chat, which grouped conversations and reference files without threads or a coordinator. In the replies they add that it reaches Pro and Max users who already run cloud sessions and have not used Projects first, earlier Projects users a little later so existing ones move over cleanly, and that support for local workflows is coming very soon. A good deal of what is built on eChai, this site included, is built in Claude Code, so this is the next thing a lot of founders here will be running their week through.
ClaudeDevs
Today we're rolling out Projects in Claude Code on desktop and web.
A project is one conversation with Claude. It splits the work into threads itself, runs them as parallel cloud sessions, passes context between them, and keeps going when you leave.
🚁 *Introducing Vayu Veer UAV Systems by RCSL — Engineered for Critical Missions.*
Proud to share *Vayu Veer*, a versatile UAV family developed to support diverse real-world operational requirements.
From infrastructure inspection and night surveillance to firefighting, emergency response, rescue operations, and critical logistics, *Vayu Veer is designed to take capability where it matters most.*
What stands out is its *mission-ready and modular approach*, built around practical operational needs.
🇮🇳 *Vayu Veer UAV System by RCSL*
Designed for critical missions. Built for impact.
The best exits don't start when the banker enters the room. They start twelve to eighteen months earlier, in the quiet work of tightening margins, cleaning cap tables, and rehearsing the narrative until it's the one buyers repeat back to you.
By the time due diligence begins, the outcome is largely priced in. What looks like negotiation is often just discovery, the buyer verifying what your positioning already told them.
At Stir Advisors we treat sell-side mandates as an operating programme, not a brochure process. [...]
Most rounds are lost in the meeting before the meeting, the informal partner call that decides whether your deal even reaches the Monday partnership.
Term sheets are downstream artifacts. If the narrative, comps, and champion are wired correctly, terms tend to normalise. If they aren't, no clever mechanic recovers the round.
Whether equity, debt, or structured capital, dilution and cost of capital are the last things to negotiate, not the first things to fear. Clarity of use of proceeds and path to liquidity always outrank valuation theatre.