What did Boris Cherny (@bcherny) post on X, week of Sep 15 – Sep 21, 2026?

@bcherny's week was dominated by shipping — Projects in Claude Code launched midweek and absorbed most of his attention, with Claude Mods, the chat/Cowork merge, and in-conversation Docs/Slides/Design rounding out a cluster of product announcements that he backed with hands-on usage details and direct user support.

Weekly brief · week of Sep 15 – Sep 21, 2026

Get this account in TelegramFollow

Product Announcements & Features

~60%

Product announcements drove most of @bcherny's week, anchored by the Projects launch in Claude Code on Sep 17 and its ripple of follow-up posts through Sep 18. The through-line: Anthropic is collapsing separate concepts — chat, Cowork, Code, docs, slides, design — into fewer surfaces, and Projects is the centerpiece where that convergence happens. [1]

Projects launch and personal workflow: The biggest thread of the week — 35 replies — was @bcherny's post that "Projects have changed not only how I interact with Claude but how I code" [1]. He says he stopped managing sessions: "I just send thoughts as they come, Claude splits them into threads, and the project remembers how I work." When asked about TDD and loops, he said Claude handles those now — "No need to tell Claude to do it anymore" [1]. He confirmed he uses Fable 5.1 at low effort for the project coordinator [1]. A separate post reinforced the same point: "Projects are how I write a lot of my code these days" [2]. He pushed back gently on the naming — "we didn't want to add yet another concept that people have to understand" and chose to evolve the existing Projects concept rather than introduce a new one [3]. When someone dismissed Projects as "folders that learned marketing," he reframed them as "folders that you can talk to, so you don't need to touch the files in them" [2]. He confirmed CLI support is coming soon and that desktop is "super good now" [1].

Chat and Cowork merging into one Claude: On Sep 16, @bcherny announced that chat and Cowork are merging into a single Claude that carries context across everything [4]. The framing: Claude Code proved AI could ship real code, Cowork proved knowledge workers could hand off briefs and get finished files, and now those modes live in one conversation. He said he's been using it daily for weeks and it feels "simpler, faster, and more powerful" [4]. A user agreed the separation never made sense; @bcherny asked for clarification when the user mentioned wanting "visual" agents like Grok [4].

Claude Docs, Slides, and Design in every conversation: Also on Sep 16, @bcherny highlighted that Docs, Slides, and Design are now embedded in every conversation — ask for a presentation and you get one you can open, edit, and export as PowerPoint or PDF, with no separate tool to navigate to [5]. When a user pressed on PowerPoint build efficiency, he said the team is "working hard to make it as efficient as possible" [5].

Claude Mods landing: The week opened Sep 14 with @bcherny announcing Claude Mods, flagging that someone already built a Tetris-in-Claude mod [6]. The 31-reply thread drew community demos — an rwx CI panel built in 20 minutes, a breathing exercise mod for focus while waiting — and he confirmed Mods work on desktop and web, not just CLI [6].

Product Support & Engagement (~10%): Beyond the launch threads, @bcherny spent time on direct support. He flagged a login-bug report from @tdinh_me as "weird" and said he was looking into it [7]. He asked a user reporting re-auth issues for their version and setup details [8]. When a desktop-app user complained about update banners, he said they've been improving this and asked if the banners persist on the latest version [9]. He also surfaced a sidebar-collapse fix by quoting @aaandeverything's reply to @mitchellh [10].

Personal / other (~30%): The rest of @bcherny's authored posts were quick reactions — praising community animations and demos ("This is extremely cool," "Extremely cute," "Wow, love the vibe"), cheering a new podcast from @kevinroose and @CaseyNewton as a "dream team," and agreeing with @rauchg on Vercel's sub-second deployments [11][12][13][14][15][16][17][18][19][20][21]. Mostly one-liners, no substantive off-thesis content.

35 replies · 16.2k viewsHis most-discussed post — 35 replies — described how Projects changed his coding: he stopped managing sessions and just sends thoughts as they come, Claude splits them into threads, and the project remembers how he works. He confirmed Claude now handles TDD and loops on its own and that he uses Fable 5.1 at low effort for the coordinator [1]. [1]

31 replies · 10.8k viewsHis 31-reply thread announced Claude Mods are landing, flagged a community Tetris mod, and drew user demos including an rwx CI panel and a breathing-focus mod. He confirmed Mods work on desktop and web too, not just CLI [6]. [6]

20 replies · 9.9k viewsHis 20-reply post announced chat and Cowork merging into one Claude that carries context across everything, with no separate mode to switch between. A user asked about "visual" agents like Grok; he asked for clarification [4]. [4]

18.1k viewsretweeted @emollick's post on Claude Projects as an agent orchestrator that spins up specialist agents — 18 parallel threads for historical-mystery research, each launching sub-agents for simulation, codebreaking, and fact-checking [22]

806.5k viewsretweeted @claudeai's announcement that Salesforce integration is in beta, bringing accounts, opportunities, and pipeline into Claude with 37 pre-built sales skills [23]

211.5k viewsretweeted @trq212's post that Claude Code now supports AGENTS.md as a fallback when no CLAUDE.md exists, starting in version 2.1.277, toggleable in /config [24]

181.6k viewsretweeted @addyosmani's post sharing Anthropic's blog on agentic coding straining CI — Claude writes 80% of code, engineers ship 8x more per quarter, tests grew 10x, and CI jobs are up 25x in 6 months [25]

53.8k viewsretweeted @felixrieseberg's post describing the Cowork+Chat merge: fire off questions then do serious work in the same conversation, with Claude using local files when your computer is open and continuing in the cloud when you close it [26]

Get this account in Telegram, free.Mirra posts every new brief to a private channel. Joining a channel that already exists costs nothing.Free to join · no X or Telegram credentialsFollow this account

How Mirra writes a brief

Three steps, the same way every time — so a brief can be checked rather than trusted.

1
Every public postThreads, replies and quotes the account published inside the cycle window. Retweets are read for context, never counted as authorship.
2
Grouped by topicMirra weighs how much of the window went where, then writes one summary per topic — the dominant topic gets subgroups.
3
Nothing unlinkedEvery claim and stated position points back to the post it came from, so you can check it in one click.

Questions

A summary of everything one X account posted inside a cycle — grouped by topic, with a citation on every claim back to the post it came from. Mirra writes it; the account does not.

Once per cycle. A weekly brief is published at the end of its seven-day window, and this page then shows the newest one.

Yes. Add any public X account and Mirra starts covering it — you get briefs in the dashboard, a private Telegram channel or by email, on the schedule you pick.

Mirra emblem

Follow the accounts that matter. Skip the feed.

The algorithm hides posts. You both lose. Mirra reads every public post, groups what was said, and delivers it on your schedule.

Start free trialBrowse public briefs

This is an automatically generated summary of @bcherny's public posts on X — the linked originals are the source of record. Mirra is not affiliated with Boris Cherny.

Are you @bcherny and want this page removed? Email support@mirra.to.