Joined April 2022
1,722 Photos and videos
Pinned Tweet
Fable 5 handoff Fable 5 is an amazing model, but expensive / slow (holy thinking tokens) Here's how I've been using Fable 5 to plan, GPT 5.5 to execute, & Composer 2.5 to explain
7
1
95
8,805
New in Conductor - in-app browser Experimental by default (settings > expirimental > in-app browser), this browser features @Agentation_ from @benjitaylor Edits are passed directly to the Composer & your harness of choice! Visual edits, just like that
8
6
88
7,139
pre-train post train matt train
6
43
8,774
New in Conductor v0.65.0: - Codex Goal support - Drag & drop reordering in the queue - Force archive workspaces Cloud beta waitlisters might want to keep an eye on their inbox, too!
21
7
127
7,001
congrats I guess?
3
20
3,694
Fable 5 handoff Fable 5 is an amazing model, but expensive / slow (holy thinking tokens) Here's how I've been using Fable 5 to plan, GPT 5.5 to execute, & Composer 2.5 to explain
7
1
95
8,805
this is how I get my steps in now
3
4
1,514
Great to see this formalized Fable 5 is really great at refactoring / optimizing PRs, but often I want to understand why Rather than asking follow ups at $10 / $50 per million tokens, I hand these off to @cursor_ai Composer 2.5 (5% of the cost) Recently, I asked Fable 5 to refactor a PR - I noticed it made a large change to a SQL statement and added a CTE I handed off a transcript of the chat to Composer and asked "Why did you change the SQL statement for the queued message updates?" Fable 5 works while the Cursor harness explains The same thing can be done with a plan implement approach (Fable 5 -> GPT 5.5) Since 5.5 is ~50% cheaper, the math still holds
Jun 10
You have Claude Fable for only a few days. Here's how to make the most of it. Introducing /improve: use your most capable model to audit your codebase and write plans for cheaper models to execute later. Studies your code, figures out bugs, perf, tech debt, missing tests, what to build and writes plans any agent can run.
10
4
89
20,601
now that’s a great looking dock
8
61
10,287
pretty good feeling when a tutorial video results in a first-party integration
Your Hermes Agent now lives in iMessage via @photon_hq. Run 'hermes gateway setup' and choose Photon to start texting your agent.
4
1
30
5,602
Swing by - will be showing how I use conductor to build my projects (and conductor!)
What are people at the top of the game building with AI agents and how are they doing it? What works for them and what doesn't? Next week @twiecki (@pymc_labs) and I will learn from - @JnBrymn, who is neck deep in agent harnesses and context engines, - @isaac_flath, who is an ai eval and content master, and - @mattyp, who's working on agent parallelization and orchestration @conductor_build. ​Register to join live or get the recording afterwards. luma.com/0t8kiodw
1
8
2,051
Claude Fable 5 is now available in Conductor conductor.build/changelog
Claude Fable 5 is now available in Conductor conductor.build/changelog
12
2
61
9,375
1
2
22
1,416
"so we're just announcing it" is amazing copy
We recently submitted a confidential S-1. We expect it to leak so we’re just announcing it. We have not decided on timing yet; it may be a while because there are things we want to do that are likely easier as a private company. But it’s a complicated set of tradeoffs and this gives us the option to go public sooner if that ends up being best. This announcement is being made pursuant to Rule 135 under the Securities Act of 1933, as amended, and does not constitute an offer to sell or the solicitation of an offer to buy any securities. Any offers, solicitations of offers to buy, or any sales of securities will be made in accordance with the registration requirements of the Securities Act.
7
54
18,715
Conductor now supports @cursor_ai with Composer 2.5 With native support for 3 harnesses, we now have the holy trinity (Claude, Codex, Cursor) and have fulfilled the prophecy of the 4C's Conductor, Claude, Codex, Cursor
We've added a new harness! Cursor Composer 2.5 is live in Conductor. It's fast, precise, and cost-efficient. And when I say fast I mean _really_ fast. Excited to hear your thoughts!
29
4
181
27,323
as a huge @raycast fan, this makes me happy
few things i wanna try - poke @interaction - conductor by @conductor_build - opencode @opencode
2
37
6,621
Always great chatting with the legend @tonykipkemboi
just dropped a new episode of @AIAgentFrontier with @mattyp, now leading developer experience at @conductor_build after his awesome run at @Replit. other than a good reason to catch up, i wanted his read on where the ui for agents are actually headed. Matt called it the iPhone moment for agent communication. we're all running agents in parallel, steering them, switching between Claude Code and Codex, and the UI layer for all of it is still being invented. that's the bet conductor is making. we also got into how devrel is quietly becoming builder relations, and why being authentically yourself is the one thing the models can't copy. full episode's linked in thread.
3
17
7,420
This is the best tool I've found for getting company logos, svgs, & brand assets >search for any company >copy svgs directly (or code for your site) >public API >raycast extension (grab svgs from your desktop) >AI friendly as of last week you can now get @conductor_build assets there as well :) check it out svgl.app/
17
13
341
30,663
What's new in 0.62.0!
Conductor 0.62.0 is out now: - File-backed settings - In-app browser, powered by Agentation - Queue improvements - Easier chat forking conductor.build/changelog?v=…
5
5
81
14,774
Live & chatting for a bit!
1
5
1,992