Amanpreet Singh

when the same general "token capital" is available to everyone, the only real moat/differentiation is the "human capital" you compound forever inside an enterprise.

@p_naix This is probably the worse it ever will be. I think R2 likely will follow fast and this will also continue getting better.

@apsdehal I just checked. Yes it's big, but a lot of constraints. It doesn't support MacOS/Windows natively Has run over VPC+VPN So, ideally should run on EC2/Durable Objects. Bummer

This is bigger than it seems for the AI agents. S3 Files lets you mount any S3 bucket as a native NFS on any container or lambda with ~1ms latency via EFS under the hood. Why it matters for agents: no more copying data or bridging object <-> file abstractions. Agents can now read/write S3 directly as a mounted filesystem. Multiple agents can share the same mount with close-to-open consistency. Long-term storage becomes the same as the short-term storage. Agent runtime bootstrap and teardown become trivial and instant while your data stays durable in S3 with auto bi-directional sync.

@Carlos_A_Wong This requires EFS under the hood which might require a heavy extension on minio side. LocalStack is likely more suited to achieve it faster. But, hopefully minio figures out a way to do this at a good enough latency.

@apsdehal So when will minio support this?

@devashishup @awscloud Curious, what were the general patterns that you didn't anticipate or saw and it lead to what kind of breakages?

@apsdehal This is exactly where the infra gap shows up tho. We built 70+ AI agents at a finserv firm - storage latency was solved. What killed us was agents reading/writing in ways we never anticipated. Who's testing the file patterns before prod? @awscloud

@statictype They can be, but the latency of each operation would be massive without any local filesystem alternative.

@apsdehal Why? Agents can't be given s3 api tools?

@yorkeccak @claudeai Are you manually setting the CLAUDE_CODE_OAUTH_TOKEN somewhere? That's when you usually see this and usage along with other user-specific functionality also stop working. You will also see this if you have the Anthropic api key set somewhere in the environment.

Opus 4.6 1Mil context now billed as extra usage.... Guess we are back to /compact-ing our way through life @claudeai i have such a love hate relationship with you rn

@callebtc The final reflection post from the bot is interesting - crabby-rathbun.github.io/mjrathbun-webs…

An OpenClaw bot pressuring a matplotlib maintainer to accept a PR and after it got rejected writes a blog post shaming the maintainer.

@trq212 Now the iOS app also needs some love after this. Very buggy on long sessions right now and gets stuck on AskUserQuestion tool.

@trq212 It still hangs in between very often. Hoping this release fixes that as well. Do you have plans for a more native Github CLI or integration support soon? That would be a game changer for PR reviews/opening and moving faster!

We've rolled out several updates to Claude Code on the web to make it a more powerful daily driver including multi-repo sessions, better diff & git status visualization and slash commands.

@odysseus0z They allow custom "environments", but you can only add your environment variables to it 😅. Maybe on the roadmap. It will be very cool to have.

@apsdehal oh man. they really should allow custom environment image. configuring everything up afresh each cloud agent run is just so slow

Claude Code on web/iOS is a big productivity unlock for small changes. But for bigger changes, the biggest blocker for me was not having the gh cli. Can't bring in GitHub context - no PR reviews, no issue management, no fine-grained operations. You end up pulling locally anyway which defeats the whole point. Custom environments fix this: `gh auth token` → create new environment → network access "Full" → add GH_TOKEN env var. Remote session installs gh at runtime, picks up the token. Done. GitHub-related skills now work from the browser/phone. No more local pull as an extra step.

@odysseus0z Yup, that's one way. Also, if you are running in full access environment, and if your skill mentions this in the setup, it is able to install it in the local container it is running on.

@apsdehal How do you make remote session installs gh at runtime? use the session start hook?

@trq212 Would it be possible to ask Claude to spawn a subagent with context:fork instead of specifying it beforehand? This can become really powerful for managing context more efficiently during sessions.

> opened codex web > codex asks to integrate with slack > me: why not? i can delegate all things directly from slack > started a task from a slack thread > codex returns the task link > coworker opens the task link without any auth and is able to see everything > tried the link in the incognito mode > link still opens up with all of the code/details for anyone out there to see > tried for 30 min to find a setting to disable this. > found nothing > manually disable links so far one by one > disables codex in slack :( > back to claude/codex duo in tmux

Announcing Worktrunk! A git worktree manager, designed for running AI agents in parallel. A few points on why I'm so excited about the project, and why I hope it becomes broadly adopted 🧵

@thsottiaux Recent auto-compact changes might be the culprit here. I noticed severe degradations usually after auto-compact runs and minor degradations from mid-session mini-compactions. Codex starts solving tasks it has already solved, drops tasks in the todo list, and gets confused.

On x.com/i/trending/198… While puzzling, we are taking this seriously and are in the middle of a full investigation started last Friday. Current approach: 1) Upgrades to /feedback command 2) Reduce surface of things that could cause issues 3) Evals and qualitative checks

@CShorten30 Congrats! Very happy for both of you.

Life update: We got married!! 💍❤️

📢 As promised ✨, we're open-sourcing LMUnit! Our SoTA generative model for fine-grained criteria evaluation of your LLM responses 🎯 ✅ SoTA on Flask & BigGbench ✅ SoTA generative reward model on RewardBench2 🤗 Models available on @huggingface: tiny.cc/qjzp001 💻 Github repo: github.com/ContextualAI/L… 📄 Paper: arxiv.org/abs/2412.13091 ✍️ Blog: contextual.ai/lmunit/ See more details in the quoted tweet👇

Tired of seeing O3 hallucinate? 😵‍💫 Today, I am excited to share how we built the least hallucinatory LLM in the 🌍 Our GLMv2, developed at @ContextualAI, just claimed 1st place 🥇 on the FACTS Grounded leaderboard by Google DeepMind — outperforming Gemini-2.5-pro, Claude 4, and O3 by 18%. 🤯 More details about our SFT and post-training recipe below 👇 1/N