From @OpenAI DevDay to @augmentcode... GPT-6.1 Sol is now available in Cosmos!
We'll be trying it in place of GPT-5.6 Sol and GPT-6 Sol in the coming days to power our software factory!
Claude Sonnet 5.5 by @AnthropicAI is now available in Cosmos!
Over the next few days we'll be rolling it into our agent loops in place of earlier Sonnet models, and tracking how it changes what our software factory ships.
GPT-6 Sol and Luna by @OpenAI are now available in Cosmos!
We'll be upgrading all our workflows from the GPT-5.6 series of models in the coming days and measuring the impact on our software factory...
I typically use GPT-5.6 Sol for my day-to-day coding, but will launch Fable 5.1 for complex bug investigations. Seeing Fable one-shot the root cause of a bug still sparks joy.
Claude Fable 5.1 is now available in our model picker! We're starting off with the same use cases as Fable 5 - extremely long, multi-step tasks that require deep reasoning - and we're increasingly pushing the model to complete tasks that can run unattended.
Try it today in Cosmos: https://t.co/zHW6QAhxiB
Yeah, yeah, another 4.5× AI productivity post 🙄
@AkshayUtture forgot to emphasize
𝗧𝗵𝗶𝘀 𝗰𝗼𝘂𝗹𝗱 𝗯𝗲 𝘆𝗼𝘂. In minutes!
1. Log in
2. Paste this link into our advisor
3. Say: "I want that"
It handles the rest. Stop overthinking it: https://t.co/G7Zhfl0rDL
Your PR sat for three days. CI was red, comments piled up — and your reviewer didn't have the confidence to hit approve.
A fleet of agents performs multiple rounds of review, fixes the CI, the comments, the conflicts, then hands your reviewer a deep review briefing and end-to-end evidence.
Humans still approve and merge 👇
Have you wondered why your single bug-fixing agent doesn't actually burn down your team's bug backlog effectively? Read this insight from Augment's CTO and co-founder @igoro, and let us know what you think!
If you’ll be in town for @TheLeadDev NYC this week, join us on Thursday for a practical, half day workshop on building your own software factory
Request a spot here: https://t.co/lkbMlit98M
This has been my dev environment for the last 6 months. Leverage the pre-built Experts or ask Cosmos to customize Experts based on your preferences. And you have to try the mobile interface!
Quick start: a software factory in Cosmos is two screens and one prompt.
1. Connect your repo
2. Create an environment
3. Describe the factory you want
Advisor picks the experts, writes the configs, applies them, and asks you when it needs a decision.
Reposting this because I rely on it every day.If I have to step away to drive my kid, I don’t lose momentum. I can keep tabs on the agent from my phone and steer it via voice notes as I go.
The best part of an agent that runs in the cloud isn't the speed. It's that you can walk away.
Cosmos notifies you the moment it needs you. Everything else, it handles.
VFS is the perfect companion to an agent.
It turns isolated runs into a shared hive mind—persisting and sharing knowledge across sessions, agents, and users.
#AIAgents#MultiAgent
Watch a PM and an engineer collaborate on a PRD — with their agents, without leaving Cosmos.
A PM drafts it with her agent. An engineer opens the same file, has his own agent review it, leaves a comment. Her agent answers it and updates the doc.
One filesystem. Both people. Both agents.
* Remind yourself that the agent will tirelessly introduce complexity into your system and it's your role to exercise good judgement to reduce that complexity.
I'd love to hear if there are other techniques that work for people. 5/5
When dealing with agent generated code, your review process must change to maintain good architectural design and minimize complexity. I have yet to find a model that make the same design tradeoffs of a good engineer. I try to focus on the following: 🧵 1/5
* If the review agent surfaces a list of potential issues, slow down and have it explain each issue to the point you can make a judgement on each issue's importance. Agents are great at over engineering, and it's easy for the reviewer to get overwhelmed with details. 4/5