Biswaroop Palit

Products at @box| Love traveling, hiking, squash, football and @FCBarcelona| Foodie| Views my own

[Biswaroop Palit @bpalit]

This is why I’m so excited about the future of @Box

Enterprise content is no longer just processed by humans. It is becoming a programmable knowledge layer that AI agents can reason over.

The opportunity is to make that intelligence available securely across every application, developer workflow, and agent while preserving permissions, governance, and trust.

That is the promise of headless Box: governed enterprise knowledge for AI agents

[Aaron Levie @levie]

The reason I have an unhealthy obsession with AI right now is because I've spent my entire professional life on essentially one problem: how do you increase the value of content in the enterprise. How do you secure it, how do you collaborate on it, how do you govern it, and how to integrate it across all your applications.

But there's been one glaring issue that we've dealt with since the founding of Box. We could never really process information at scale in any real automated way. There have been many attempts at this problem (often in the search space), but nothing that really fundamentally transformed what you can do with enterprise knowledge.

For years the primary kind of data that we could query, analyze, and process with computers was structured data. This meant anything you could shove into a database you could understand with computers - your CRM, ERP, product analytics, HR, and other data.

But all of the unstructured data that powers our daily knowledge work - marketing assets, contracts, financial documents, medical research, engineering documentation - was only valuable when a human was operating on it. There was just simply no real way to apply automation at scale to any of this data, which meant all knowledge work was largely rate limited by our ability to process information ourselves, often manually.

AI models have obviously dramatically changed this reality. And the past couple weeks perfectly highlight this incredible progress. GPT-5.6, Fable 5, Grok 4.5, Muse Spark 1.1, and a leading array of open weights models are all showing incredible advancements on working with unstructured data.

The inherent broad intelligence, reasoning, math, and coding skills in these models, combined with deep domain expertise trained into them across finance, legal, healthcare, life sciences, and other critical fields, means that we're able to completely change what we can do with this unstructured data at scale.

What this unlocks is the ability to ask insanely complex questions of your data that were never before possible, and let agents just run on for minutes or hours across these data sets to accelerate knowledge work.

And it's not just about automating the work that we already do. While this is highly valuable, it wouldn't be particularly transformative. What's exciting is that you can now throw compute at unstructured data problems that wouldn't have been possible before. Analyze every risk on my contracts, do due diligence more deeply on a prospective investment or acquisition, look through all past client interactions in an industry to find best practices to replicate, comb through life sciences research or clinical trial data for new insights, and on and on.

So that's why we're insanely excited about what AI Agents can now do with content on Box.

[Biswaroop Palit @bpalit]

Great example of the cutting edge work you get to do on the @Box DevRel team.

We are hiring for more - https://t.co/gi8zGGYv6q

[Elon Musk @elonmusk]

Neat

[Biswaroop Palit @bpalit]

@ben_m_somers @reedhastings what student age group / grades will you target for this?

[Biswaroop Palit @bpalit]

@_philschmid @Box Will do - we plan to test out some long running tasks as well

[Biswaroop Palit @bpalit]

We have seen many enterprise customers build out internal agents that need access to enterprise content. These agents need access to enterprise content.

@Box MCP server is quick to get started with and reduces friction for developers to provide governed access to enterprise content. Great to see it in action with Gemini Managed Agents and the support for long running tasks!

Managed agents need managed content :)

[Logan Kilpatrick @OfficialLoganK]

Today we are rolling out some big updates to Managed Agents in the Gemini API:

- support for background tasks
- remote MCP & function calling
- network credential refresh

and you can now get started with Managed agents in the API via the free tier!

https://t.co/lc7giVtekq

[Biswaroop Palit @bpalit]

@DynamicWebPaige Love it! We just did some initial testing at @Box https://t.co/29XhGJpdHf

[Box @Box]

AI agents need more than intelligence. They need trusted business context.

With @GeminiApp Managed Agents now supporting remote MCP servers, developers can connect Gemini agents to governed content in Box, generate outputs from real enterprise files, and save the work back where teams already collaborate.

Box MCP Server + Gemini API in action.👇

[Biswaroop Palit @bpalit]

@FabrizioRomano Always better to let your game do the talking ;)

[Biswaroop Palit @bpalit]

@rashmi Where are the games broadcasted in downtown MV?

[Biswaroop Palit @bpalit]

As a life long Argentina fan - nothing but respect for Cape Verde! This World Cup is going to be GDP changing for them - what a way to announce themselves to the world

[Connell McShane @connellmcshane]

If you watch that and still don’t like soccer, then you never will. The best player of all time vs. one of the biggest underdogs of all time produces one of the best games of all time. Unbelievable

[Kalshi FC @KalshiFC]

🚨 Vozinha has completed more dribbles at the Knockout round of the 2026 World Cup than Cristiano Ronaldo.

[Biswaroop Palit @bpalit]

@NealGardner_ @SEFutbol needs to up their pregame to ensure fair referees

[Biswaroop Palit @bpalit]

@petergyang @NousResearch @karan4d Would be interesting to hear if teams use Hermes vs just individual setups. If yes how is that orchestrated

[Biswaroop Palit @bpalit]

@lennysan post sponsored by @OpenRouter ? IYKYK

[Biswaroop Palit @bpalit]

@hwchase17 s “wiki memory” really resonates with how we’ve been thinking about Enterprise Knowledgebases.

Agent memory is still early, but one pattern we observe consistently: enterprises don’t need agents to retrieve raw documents, Slack threads or tickets at query time. They need to turn all of that messy source material into a durable, compact, inspectable knowledge layer that future agents can rely on.

In the enterprise, memory cannot just be a folder of markdown files sitting somewhere. It needs to be governed, permission-aware, shareable across team members etc. This is where @Box plays a critical role.

Box already sits on top of a huge amount of enterprise knowledge. The next step is making that content agent-ready: synthesize the raw source material into structured knowledge, keep it fresh as the underlying content changes, respect permissions etc.

That turns Box from a system of record for enterprise content into a governed memory layer for enterprise agents. We wrote about this here https://t.co/ts3OpJVYdO

The future “Company Brain” in the enterprise will be built from trusted source content, agent-maintained synthesis, and enterprise-grade governance.

[Harrison Chase @hwchase17]

https://t.co/SsJz6y2pBy

[Biswaroop Palit @bpalit]

Come for the t-shirts - stay for spicy enterprise AI takes!

[Chris Kim @chriskim_dev]

If you are coming to @aiDotEngineer SF, I want to talk to you 🙂

Come say hi to the @Box booth!

Lets talk MCP, agent content layer, RAG, and agent multiplayer

[Biswaroop Palit @bpalit]

When building AI agents for the enterprise, your hardest problem isn’t the model. It’s getting agents access to the right content, with the right permissions, safely. Now developers can start faster with: npm install box

[Box @Box]

Today we’re shipping a new, easy way to build agents & apps with Box:

`npm install box`

This new NPM package bundles the Box Node SDK and Box CLI, so that developers (and their agents!) can rapidly scaffold, prototype, configure and bootstrap the intelligent, content-powered experiences they are building.

https://t.co/FSrRG36q70

[Biswaroop Palit @bpalit]

Enterprise knowledge bases that are governed and human curated are becoming very common as AI is rolled out in the enterprise.

Here’s how we have seen customers implement them - knowledge base is kept updated via AI with a human gate keeper

[Box @Box]

If you're building agentic enterprise workflows, you've probably hit this wall: agents need trusted knowledge to act on, but most enterprise content is scattered, ungoverned, and impossible to verify at scale.

We break down how to solve it using Box as a governed knowledge layer:
→ Claude connects to Box through the Box MCP server to review uploaded content, extract metadata, and flag issues like stale dates or missing owners
→ Box Automate routes it through a human approval workflow
→ Once approved, it lands in a curated Box Hub where agents answer questions from Slack and cite the source files.

Simple as that. Get the step-by-step here. 👉 https://t.co/UF50E7JDMt

[Biswaroop Palit @bpalit]

@swyx @aiDotEngineer @brendanhunting @TedLasso @USMNT @philipkiely Amazing! Do all sponsors get a ticket?

[Biswaroop Palit @bpalit]

The CIO mandate for the agent era: don't let every team build its own isolated agent knowledge layer.

Models change, interfaces vary but every enterprise needs a trusted knowledge layer - curated, permission-aware, versioned, governed - that any agent can read and cite.

[Box @Box]

https://t.co/XKRyqndvm8