I built a @NousResearch Hermes workflow that my team at @Box uses to track AI trends. Every day at 7am, it researches what's happening across the AI ecosystem and generates a brief for our team. We use it to identify: • AI trends worth paying attention to • Open source projects gaining momentum • New tools developers are adopting • Content opportunities to work on next I've been building agentic workflows for months, and this is one I've actually integrated into my team's daily workflow. Here's how I built it:

Had a blast last night at @digitalocean's developer meetup in SF. I think sandbox agents will become the backbone of many automations teams run. DigitalOcean Managed Agents is thoughtfully designed for that. Its Action Gateway keeps credentials separate from the sandbox, so that the agent never sees your secrets. I demoed how Box Mount mounts @Box folders into a sandbox microVM, so agents can work with cloud files like local files. It’s fast and cheap to access those files, and you can mount or unmount them on demand as the sandbox comes and goes. It was exciting to see how these innovations are coming together to change how we work with our unstructured content with AI agents! Thank you for having me DigitalOcean team!

Pi 1.0 is out

Sandbox AI. Long running agents. AI context layer. SF. Join us today if ☝️ is you. I'll show how you can use Box Mount to mount a folder on @Box to your sandbox agent in @digitalocean's Managed Agent so your agent can work with your files on cloud like its in your local environment. Super excited to be speaking here today!

I hopped on two calls with @Box customers today who are using the Box developer platform to automate knowledge work. Hearing how they're turning days or even weeks of work into a couple of minutes made me so happy to be working at a company that continues to improve the customer experience by adapting to the agentic world we're living in. In this video, I show how I automate a broker-dealer audit workflow end-to-end by connecting @ChatGPT Codex to Box using Box CLI. Full video in the post below if you're interested

Give your coding agent the security policy it needs from Box. Our tutorial uses @Vercel fx and the Box CLI to retrieve, apply, test, and upload release evidence back to Box with your approval.

At Box, we've been testing Sonnet 5.5 in early access on our complex work eval with the Box Agent, and Sonnet 5.5 is another strong jump from Sonnet 5 on knowledge work with enterprise content. Across the board we saw a 4 pt overall improvement on our hardest tests, but it was roughly 2.4× faster to a finished deliverable on 12% fewer tokens. We also saw major wins in key verticals, like a +18 percentage point improvement in Financial Services, +7 pt in Legal, +8 pt in Life Sciences, and +7 pt in the Public Sector. Here are some specific wins on certain tasks that show the power of Sonnet 5.5: * Financial services (+48 points): a due-diligence review of a trading-tool acquisition, where the deal book's arithmetic doesn't hold up. Sonnet 5.5 caught the miscalculated interest totals and the mispriced options and flagged them, and it also finished this work about 61% faster. * Legal (+14 points): a commercial lease review. Sonnet 5.5 refused to invent a "standard" market benchmark for renewal terms and notice periods, and instead flagged what the lease was genuinely missing. * Life sciences (+22 points): writing up a field trial from the raw measurement data. Sonnet 5.5 got the sample standard deviations right across both treatment groups. Sonnet 5 missed nearly all of them. And Sonnet 5.5 did it about 41% faster. * Public sector (+11 points): a report on a math intervention program. Sonnet 5.5 pulled the right student progress-monitoring figures out of the underlying data, about 71% faster. Customers will be able to build AI agents with Sonnet 5.5 in the Box AI Studio shortly.

When your cloud file storage is AI-native, coding agents can automate file-heavy workflows across hundreds of documents. See how connecting @Box to @ChatGPT codex can automate broker-dealer policy audit report prep!

Agentic orchestration, governed content, structured extraction, auditable logic, and human approval in one flow. Check out how @LangChain and Box make that possible. 👇

Great turnout at the @Box booth at @WeAreDevs this afternoon! Everyone was definitely here to learn about AI agents and filesystems. The raffle was purely coincidental 😬

20mins till the Xbox raffle devs at @WeAreDevs! Come sign up NOW before you’re too late

Devs at @WeAreDevs! Come to the @Docker booth at 2:30pm today! learn how you can mount your company files/folders in Docker sandbox so that your long running cloud agents can work with your company knowledge!

I’ll be showing long running cloud agents in @digitalocean managed agents that has access to files on @Box to fully automate agentic workflows in day to day work!

Set your alarm! ⏰ @WeAreDevs Box booth @ 1pm We're raffling off an Xbox and will use @claudeai to build an app on the fly that picks the winner. You must be present to win, see you there!

We gave Meta’s Muse Spark 1.3 a product launch to rescue: 7,000 units due in one week after a supplier shutdown. In this Box AI demo, one prompt sends it through 24 documents to reconcile conflicting inventory records and reject an incompatible substitute. It generates the lowest-cost feasible recovery plan: $103,700 to meet every shipment deadline, with a cited daily schedule and the approvals needed before anyone buys materials or books freight. Muse Spark 1.3 is now available in Box AI.

Muse Spark 1.3 is now available in Box AI. What our evaluation of Muse Spark 1.3 found: → 13% accuracy lead over comparably priced models on the Box Complex Work Eval → 42% faster than Muse Spark 1.2 using roughly a third fewer tokens → Strongest gains in Financial Services, Technology, and Energy

Box devrel team is at @WeAreDevs! Come learn how to give your @ChatGPT @bot @Muse @claudeai @NousResearch Hermes company knowledge with @Box as the file system! We got live demos for ya 😎

Gotta say, the partnerships team cooked with this one

Seems like @Muse is enterprise ready right out of the @Box 😉

On September 29, join us for a session with @daytonaio on how to give AI agents a real computer and the enterprise content to work on. What you'll learn: → How to think about the right use cases for sandboxes vs. legacy infrastructure → How to configure and operate Daytona composable computers → How to connect enterprise content to your agents using Box Mount If you're building agents that need to do real work on real content, you won't want to miss this. Register here. events.zoom.us/ev/Amkp36LSSk…

You can now control a remote machine where a Hermes agent is running. This removes so much friction that comes from ssh, vps setups