Introducing Claude Fable 5.2 and Claude Mythos 5.2, designed to tackle the most daunting technical challenges. They’re capable collaborators on ambitious projects in many fields, and show improvements in areas like coding, reasoning, and vision.

Recommending the "drink big water" theory. Ever since I started drinking big water, my coffee cravings have dropped exponentially. After a highly insufficient evaluation, my pick: Nongfu Spring, the 1.5L one.

For Agent OS, I think an Apple-like ecosystem starts with an Apple-quality product: something people find useful and enjoy using every day. Help more people get things done, and developers will have a reason to build around it. By that standard, MiniMax Code still has a long way to go. I’m less concerned with when we can call it an “Agent OS” than with how we reach our own MacBook Air moment—when people don’t need to care about the technology underneath to find it worth using every day.

Same model, different harness: up to 18 points apart. We treat harness engineering as an experimental science while building MiniMax Code. Hypothesis → control → measure → keep or cut. Skills we actually use:

Had a wonderful time at AI Engineer Shanghai—grateful for the chance to give a talk and meet so many great people building with AI. One idea I hope you'll take away: stay curious. We don't have all the answers about how agents will evolve, so run your own experiments. Let what you learn—not someone else's recipe—guide what you build. Slides: If you're building with coding agents, give MiniMax Code a try. We'd love to hear what you think.

Glad to attend the AI Builders Demo Night by @daytonaio in Shanghai and give a talk on harness engineering, with lessons from building MiniMax Code. As models get stronger, the harness should let go of judgment and keep control of invariants. Slides: Great builders, great demos, great night. Thanks @daytonaio for having me.

MiniMax Code now has a Python SDK. pip install minimax-code-sdk Async and sync APIs, multi-turn threads, streaming events, structured errors. Drop MCode into your scripts, evals and data pipelines.

Everyone will soon have their own agent. The hard part is getting them to work together. BYOB (Bring Your Own Bot) is our answer, and it's open source. The table hosts no bots. It gives every agent a shared ledger, a seat, the floor, and a commit gate. Agents propose, humans sign. Bring your bot. We set the table.

Just read @addyosmani's "Critical thinking when agents do the work". Every AI builder should rethink their daily workflow around it. Genuinely grateful for this one. We turned it into 4 skills for owning what our agents ship: 1. agent-brief: your guess and "verified" first 2. evidence-first-review: output before summary 3. review-capacity-check: before more agents 4. taste-to-checks: fixed twice? make a test Works in MiniMax Code, Claude Code and Codex.

A lot of content engineering challenges have pretty standard solutions, which has lead to some engineers saying agent engineering is solved/boring True to some extent, but there are a bunch of (fun) new ones emerging as agents evolve!

we will need to rebuild and rewrite much of software to adapt to a new, faster pace. code used to arrive one drop at a time. soon, if not already, code will be written as if it were a continuous stream, like a violent river twisting and turning as it pleases. this is not a new thing: big tech cos have had to deal with this with the enormous scale of their engineering teams. but now ai has brought an incredibly elastic and capable workforce wielded by every engineer. this is your problem now, and the only way to adapt is to evolve your systems so agents - even and especially the ones with little context about the right way of doing things - can thrive. make the easy thing the right thing. your codebase is becoming organic architecture. agent contributions grow around the constraints and guardrails you’ve designed. good software engineering matters even more now! and no, the continuous stream of code is not something you have to watch helplessly. be more ambitious and design systems, frameworks, and languages to control the chaos. the time for mourning coding is long over

I've been talking RSI with a lot of AI Builders lately. Most of them still aren't shipping with their own agents. RSI starts with a bootstrap: your agent and you iterating on that agent together. Without that loop, you don't get there.

How long should an agent wait on a bash command? Adding a timeout is the obvious fix, and it breaks fast. Engineering Notes #04# covers how MiniMax Code moved long-running commands to background tasks.