Devin

@jaredpalmerSep 23

Been using @DevinAI to do autoresearch on Kev overnight with @Modal Outposts (this gives Devin’s Cloud agent full access to GPUs) as well as Devin’s macOS VMs to do some work on MLX. The workflow has been super powerful. I also set up a bunch of Devin Automations for a lot of routine repo maintenance work. Definitely some room for improvement though. Need to probably tweak some automations to score PRs and Issues before just blindly taking action and trying to repro etc. same thing for Devin AutoReview. Fwiw I don’t think these headwinds are unique to us, OSS just has different security and trust model for agents vs. internal work.

@rhyssullivanSep 22

How in the world does T3 code have a mobile app and Devin doesn’t

@dabit3Sep 22

Known affiliation

Cognition is giving away 50 $200 Devin Max plans to celebrate the new model launches! ⚡ Now available in Devin: • GPT-6 Astra, Sol, Luna + others • Claude Opus 5.5, Fable 5.1 + others • SWE-2 (Free until October 15) • Fusion Frontier harness (Fable, Astra, Sol, Opus) • Gemini 3.8 Flash + others • Grok 4.7 + others • Kimi K3 + others • Inkling • DeepSeek V4.1 Flash + others • GLM-5.3 Flash + others • Cloud agents on Linux, macOS, and Windows To be eligible, reply below with what you're building (or something you'd like to build with Devin)! We will choose winners in 24 hours.

@cognitionSep 21

Known affiliation

Introducing Devin Cloud in Terminal and devin ssh: two new ways to use Devin's computer. Create, steer, and resume Devin Cloud sessions right in your CLI with /cloud. And for the first time, SSH into Devin's dedicated VM, then /handoff the work back to your own device.

@catalinmpitSep 21

The difference between Devin SWE-2 and Fable 5.1's outputs is crazy. SWE-2 is chill and writes as close to a human as possible for an AI. Fable 5.1 writes like a nerd trying to sound smarter than he actually is, which leads to unnecessarily convoluted output. I have to constantly run the /unslop skill to make the text readable. It's incredibly tiring reading such text every day.

@thekitzeSep 18

$200/mo for openai $200/mo for claude $200/mo for devin $0.5/mo for jev $300/mo for grok someone who is good at the economy pls help me budget this my family is dying

@jaredpalmerSep 16

Code Scans are a new primitive in @DevinAI. Borne out of Security Scans, Code Scans are a generalized way to deeply audit and review entire codebases against a target goal. We've made it easy to scan for stuff (like SEO/AEO issues, a11y, performance issues, memory leaks, unoptimized db queries, vulnerabilites, dead flags, design system adherence, etc.), quickly review findings, and then send Devin off to fix the ones you want. Since scans happen within a regular Devin sessions or within an Automation, you can configure/enrich your scans with MCPs, Skills, models, budget limits, and network policies too.

@cognitionSep 16

Known affiliation

Some engineering tasks require reasoning across the codebase: Which code is safe to delete? What queries are slowing performance? Introducing Code Scans: codebase-wide audits for any goal. Devin investigates, reports findings, and opens the PRs. Powered by Agentic MapReduce.

@dabit3Sep 15

Known affiliation

Devin now has a Mac  Here are 11 experiments I've run over the past week, as well as open source code for over 100 more native apps. Build, test, deploy iOS, iPad, and macOS apps in the cloud with just a prompt. ⚡ 1. Build and test a multi-platform, multiplayer app in the cloud (iOS, iPad, web)

@dabit3Sep 14

Known affiliation

much of whatever success I've had in my career can be attributed to being terminally online, and being intentional about the people and places where I consume information being able to identify and act on the right information as quickly as possible opens up a lot of weird and interesting opportunities that are hard to explain sometimes the next phase of this is having agents surface this information more quickly and precisely. automations allow me to process a vastly larger set of inputs than I could as a human for me this is looking like daily reports from Devin and Hermes agent using xAI, Exa, GitHub and other APIs that process signals I'm looking for, summarize and brief me on things I should be looking into I'm still only scratching the surface here but I am already floored by the value these types of automations provide me vs doing it all myself

@dabit3Sep 11

Known affiliation

Fusion is the most efficient frontier harness for GPT 6 Astra and Claude Fable 5.1. You can also configure everything about it: • Base model: Fable, Astra, Sol, or Opus • Sidekick model: SWE (Free), Luna, Sol, GLM (Free) • Speed: Normal or Fast ⚡ • Reasoning level It retains Claude Fable 5.1 and GPT-6 Astra performance while reducing costs and is the first time a multi-model coding agent has been included on the @ArtificialAnlys Coding Agent Index! Importantly: it's also fun to use :)

@cognitionSep 11

Known affiliation

Introducing Fusion in Devin CLI The most efficient frontier harness for Fable & Astra; 39% cheaper across coding benchmarks. Pick your favorite model for planning and a cost-effective model for execution.

@ryancarsonSep 10

The Cognition team gave me 50 free @DevinAI Max plans to give away. Each one is worth $200. Reply with the most useful or ambitious idea or project you’d have Devin ship for you this week. We’ll pick 50 people and DM the codes!

@cognitionSep 10

Known affiliation

Introducing Devin Voice 🦦 ☎️ Your favorite AI software engineer just got a landline. You say it, Devin ships it. Powered by GPT-Live and our new SWE-2 model.

@dabit3Sep 10

Known affiliation

Introducing SWE-2, Cognition's best coding model yet. Within one point of Fable 5.1 on FrontierCode at 64% less cost. SWE-2 is free in Devin CLI and Devin Desktop, 75% off in Devin Cloud. ⚡ It's tailor made for real software engineering work. Try it free with any subscription.

@dabit3Sep 10

Known affiliation

Automated UI testing feels like AGI in Devin 10 examples (running macOS + Linux Cloud Agents) 🧵 What happens (automatic for any update): • Creates programmatic test plan • Boots the app with secrets & credentials (auth, etc.) • Autonomously clicks, types, and checks the UI like a human using a computer would • Records a test • Edits the video, attaches step by step walkthrough of the actions it took • Sends you a concise video artifact As we move away from reading every line of code, new advanced ways of testing become more valuable. 1. Testing a platform for signing documents electronically

@a16zSep 9

Vals AI CEO Rayan Krishnan on the unlimited tokens experiment that revealed an anxious obligation for engineers to use models all the time, everywhere: "I wanted to do a tokenmaxxing experiment. I was able to get unlimited access for our team for a month for some of the coding tools. We had a lot of engineers spending between one to two billion tokens a day." "I went back and did some math, and it looked like in that month, we spent roughly $1.5 million worth of tokens... It was actually 10x more we were spending in tokens than employee salary for that month." "We cannot continue with this mode of operation for the next month. How do we intelligently figure out what are the right tools we should use, for what teams, and what projects?" "We found some pretty surprising insights. The Cognition Devin tool is actually very token-efficient, and so that's a place we've chosen to adopt more." "There are a lot of places where you can get better pricing models out of subscriptions as opposed to token-based pricing." "It's actually informed our strategy for how we can effectively tokenmaxx without spending $1.5 million per month." @RayanKrishnan @JenniferHli

@scottwu46Sep 8

Known affiliation

This one feels special - a lot of folks on this list we've admired for years and now finally get to work with. Above all I'm grateful for the incredible team that we have. Every day at Cog still feels like getting in a room together and trying to build something great. Also, grateful that Devin is actually good now. Would have looked really dumb otherwise.

@dabit3Sep 8

Known affiliation

Devin has been building cloud agents for over 2 years (since March 2024). In hindsight it really is crazy how far ahead of their time they were. Over the past few months (2 years later), everyone seems to be realizing that this is the best, most scalable agentic approach for building software.

@dabit3Sep 7

Known affiliation

Code Channels are now live in Slack! How to use them as a multiplayer IDE: in this demo we 1. Create a new Code Channel 2. Build a new feature 3. Invite a teammate to collaborate and review the PR 4. Receive an edited screen recording of the new feature being built and tested with Computer Use in @DevinAI Code channels are ephemeral, so once the feature is complete we remove the channel and move on.