❝

It's Wednesday, November 11th. Claude now builds whiteboards, spreadsheets and data stories in the same chat, and Telegram lets bots send small interactive apps in a private message. Plus a simple way to make a website easier for agents to use.

💡Help Shape The Collective

What should we cover next? Answer these three quick questions to help us choose the topics and technical depth you find most useful.

How would you describe your technical ability?

Login or Subscribe to participate

In Today’s Top Tools, we spotlight some of the most innovative, creative AI apps that we recommend adding to your stack.

TOOL SPOTLIGHT

1️⃣ Claude Puts Whiteboards, Sheets, And Data Stories In The Chat

Three product cards for Claude Whiteboards, Claude Sheets and Claude Data, from Anthropic's announcement video

Image from Anthropic

❝

Claude Whiteboards and Sheets are rolling out in beta on paid plans, and Claude Data is generally available on those plans. You can plan on a shared board, build a spreadsheet, and turn connected data into dashboards or stories, on web, desktop and mobile.

About the Tool

Anthropic started rolling out Claude Whiteboards, Sheets and Data yesterday on paid Claude plans. Whiteboards and Sheets are in beta. Data, previously called Claude Dashboards, is generally available, while certified queries and live sharing stay in beta.

  • Build a spreadsheet with tabs, formulas, formatting and drop-downs, then export it as xlsx or CSV.

  • Plan work on a shared whiteboard, add a timeline with owners and dates, and turn the board into a document or a deck.

  • Turn connected company data into dashboards or narrated data stories, then share a live view (live sharing is still in beta).

Whiteboards and Sheets are in beta on all paid Claude plans, on web, desktop and mobile. Team workspaces get all three on by default. Enterprise workspaces keep Whiteboards and Sheets off until an admin turns them on, and Data turns on by default on November 19, 2026.

❝

If you already live in Claude for writing or research, this is the first time the same chat can hold the plan, the tracker and the story you tell from the data. I would start with a sheet you already keep, like a hiring tracker or a budget, and see whether Claude can rebuild it and keep the formulas straight. Whiteboards look useful for a small team planning a launch, as long as everyone is already on a paid Claude plan. Enterprise admins should note that Whiteboards and Sheets stay off until they are turned on, and Data's default-on date is November 19.

Our Perspective

TOOL SPOTLIGHT

2️⃣ Telegram Lets Bots Send Interactive Apps Inside A Private Chat

A Telegram phone screen showing an Aurora Headphones widget inside a chat, from the official announcement video

Image from Telegram

❝

Telegram Bot API 10.4 lets a bot send a small HTML app inside a private chat. The widget is at most 4 KB after compression, runs on the device with no network access, and can talk to the bot or the TON wallet. Groups and forwards are test-server only.

About the Tool

Telegram added HTML widgets in Bot API 10.4, dated November 9. A bot can attach a small HTML app to a message in a private chat with that bot, as covered in Telegram's Bot News post.

  • Send an interactive widget, such as a product card or a checkout step, in a private chat with your bot.

  • Keep the HTML at or under 32 KB, and under 4 KB after zlib compression, or Telegram will reject it.

  • Call the bot or the built-in TON wallet from the widget; the sandbox has no network access of its own.

Available now in production for private chats with the bot. Widgets cannot be forwarded, and group chats, inline mode and forwarded messages are documented as test-server only. Telegram is free to use in the US.

❝

The official docs are narrower than the announcement post. Production support is for private chats with the bot, which is still a useful surface for a US customer flow, like confirming an order or showing a status card. I would not plan a group-chat or viral-forward experience on it yet, because those paths are still on the test server, and a widget that cannot be forwarded will not spread on its own. The size cap is tight, so design the UI first at 4 KB compressed. Treat TON wallet hooks as optional, and test the private-chat path end to end before you ship.

Our Perspective

🎓AIC AUTOMATIONS

🎬 How Do I Make A Website Easier For Agents To Use?

Today I'm sharing how I made a website easier for AI agents to use, with a short llms.txt file at the root and a public API. I did this on Route Doctor that I built, a free OpenRouter helper that checks whether a request will route and diagnoses errors with cited sources.

The result is live: llms.txt. It tells an agent what the site does and how to call two endpoints, checkup and diagnose. In a test recorded in the README, a fresh Claude session given only the site URL read llms.txt, called both endpoints, and gave the right fix.

Here's the pattern I used:

  • Put a plain-text file at https://yoursite.com/llms.txt.

  • In a few short lines, say what the site does, name the 2 or 3 most useful pages, and if you have an API, show one example request.

  • Keep it short. Mine points to two API paths and leaves the rest out.

I did not add an MCP server, and I did not publish Markdown copies of the pages. Those are useful later steps if you want agents to plug in more deeply. The API plus llms.txt was enough.

If you run a Shopify or WooCommerce store, try agent-readiness checker as it will also flag a missing llms.txt. It needs Node.js 22 and no packages to install: npm run scan -- mystore.com. Otherwise, use the prompt below.

Paste this prompt into Claude or ChatGPT, and swap in your URL:

❝

Here's my website: [your URL]. Write a short llms.txt I can put at the root. In plain language, say what the site does, list the 3 most useful pages, and if I have a public API, show one example request an agent can try. Then fetch [your URL]/llms.txt, which may 404 today, and tell me what to fix first so an agent could use the site.

Pro tip: Keep llms.txt short and point to one or two actions, like a single API path. A long welcome note is harder for an agent to follow than two clear endpoints.

One way to put AI to work, every week. From Joy Dong

The AI Collective Podcast, live Wednesday November 11 at 9 AM PT with Yiying Zhang of GenseeAI and Amol Hardikar of Humanoid

Live today: GenseeAI and Humanoid

🎙️ Live today at 9 AM PT on The AI Collective Podcast

Join hosts AJ Green and Catherine McMillan live at 9 AM PT today for the AI news that matters and two builder conversations.

  • 9:15 AM PT · Yiying Zhang, Co-Founder and CEO of GenseeAI, joins us.

  • 9:45 AM PT · Amol Hardikar, CFO of Humanoid, which raised a $152M Series A at a $1.35B valuation in July.

In this section, we feature a few standout opportunities at leading AI companies, non-profits, policy groups, and other organizations.

❝

🔬 Work at a Lab/AI Safety

Roles at frontier labs, safety organizations, and research institutions

RESEARCH ENGINEER, CYBERSECURITY RL - Anthropic | San Francisco, CA or New York City, NY | $500,000-$850,000 USD annual

RESEARCH ENGINEER, CONTENT UNDERSTANDING - Exa | San Francisco, CA | $180,000-$350,000

SENIOR/STAFF SYSTEM RESEARCH ENGINEER, LLM INFERENCE OPTIMIZATION - Snowflake | Bellevue, WA | $236,000-$330,000

❝

🚀 Work in Industry

Roles at funded AI startups and private companies

STAFF PRODUCT MANAGER, AI & DATA FOUNDATIONS - Plaid | San Francisco, CA | $207,600-$339,600 + Equity

SENIOR MACHINE LEARNING ENGINEER (CLINICAL TEAM) - Midjourney | San Francisco Bay Area (Hybrid) | Estimated base salary $237,000-$250,000

FORWARD DEPLOYED STRATEGIST - Magic | New York City | $100,000-$150,000 + Equity

❝

🏛️ Work in AI Policy / Governance

Roles in government, think tanks, NGOs, and policy organizations

US EXECUTIVE BRANCH POLICY & INTERNATIONAL ENGAGEMENT LEAD - Scale AI | Washington, DC | $148,000-$185,000

POLICY ANALYST / SENIOR POLICY ANALYST - AI Policy Network | Washington, DC | $80,000-$135,000 (Policy Analyst); $140,000-$190,000 (Senior Policy Analyst)

POLICY COUNSEL / ASSOCIATE POLICY COUNSEL - Encode AI | Washington, DC (Hybrid) | $130,000-$235,000 depending on level and experience

🫵 Want your message in front of 500,000 AI builders?

Our partners and sponsors get exclusive placements across the newsletter and access to AIC's in-person network — demo nights, dinners, hackathons, and forums across 180+ chapters.

For all inquiries, send us a note at [email protected].

The AI Collective is built by volunteers across 180+ chapters in 40 countries.

Thank you to the thousands of volunteers around the world who make this work possible. We truly could not do this without you.

🧑‍💻 About the Editors

About Joy Dong

Joy is a news editor, writer, and entrepreneur at the intersection of AI and blockchain. Whether she is demystifying complex systems in her newsletter TEA. Joy’s mission is to make emerging tech accessible and actionable for everyone.

Lindsay is an AI engineer, researcher, and writer focused on how AI systems behave in practice and what it takes to make them safe. Her work sits at the intersection of AI safety, governance, and product design. At AIC and in her newsletter, Hidden Layer, she writes about the questions that matter most as these systems scale.

Add Your Thoughts

Avatar

or to participate

Keep Reading

View more