Shahar Abramov

Founding Engineer @ | Maintainer FastAPI-MCP

[Shahar Abramov @ThatsShahar]

@ToriSeidenstein @ElenaVerna @clairevo looking forward to it:)

[Shahar Abramov @ThatsShahar]

@SidChoudhury @autohealai Congrats guys!

[Shahar Abramov @ThatsShahar]

@aaliya_va thanks!

[Shahar Abramov @ThatsShahar]

Tadata is live on Product Hunt today. Come support us with an upvote/comment ❤️
https://t.co/CxF4SYHz7X

[Miki Makhlevich @makhlevich]

tl;dr - Tadata is live on Product Hunt today.
Come support us with an upvote/comment ❤️
https://t.co/mPyO9LYXtF

Some thoughts on why we’re launching now, and why the world apparently needed yet another AI employee in Slack 👇

[Tori Seidenstein @ToriSeidenstein]

Recently @tadata_team sponsored our first hackathon and the first place team blew us away.

They created Arrow AI, which equips your computer cursor with AI access -- an LLM plus all your SaaS systems via Tadata-powered MCP connectors. So while you're writing an email, or typing in terminal, you can simply activate it via a hotkey and send the text to the llm, access your MCPs, and continue in context.

In a time when AI setups and data access is becoming increasingly fragmented, Arrow brings it all to a single place.

Congrats Hrishikesh Athreya, Arnav Dewan, Rohan Hareesh, and Thilak Shekhar Shriyan!

[CopilotKit🪁 @CopilotKit]

Introducing the Vibe Coding Server

An expert MCP server for vibe coding projects with CopilotKit - powered by @tadata_team

The Vibe Coding Server gives coding tools structured access to CopilotKit docs and real code snippets, automatically fetching only the relevant parts.

That means better context, fewer tokens, faster runs, and no need to tag docs in every prompt.

Just ask Cursor to build a Frontend for your AG-UI agent, and watch the magic 🪄

Links Below 👇🔗

[Tori Seidenstein @ToriSeidenstein]

Good news: we just made “vibecodability” a real metric. 📊

Better news: @CopilotKit + Tadata MCP server = 66% faster integrations.

Best news: thousands of devs already vibing.

Case study in thread 👇

[Tori Seidenstein @ToriSeidenstein]

LLMs can now read the fine print.

Even the scanned, low-res, sideways kind.

@fileAI_'s new MCP server (built w/ Tadata) turns OCR + doc AI into native agent tools.

From eKYC to insurance claims, agents can finally handle paperwork like grownups.

Full case study below 👇

[Tori Seidenstein @ToriSeidenstein]

🗝️Think you can escape an AI Escape Room game?

We built a virtual escape room powered by an MCP game engine.

Do your best to chat and (hopefully) escape to freedom!

[Tori Seidenstein @ToriSeidenstein]

FastAPI-MCP is trending on Github today 🤩

Huge thanks to everyone who’s contributed. The MCP ecosystem is growing fast!

[Tori Seidenstein @ToriSeidenstein]

Excited to announce Tadata's partnership with @tavilyai!

We’ve launched an MCP server that helps developers integrate Tavily’s API faster, smarter & with built-in AI guidance. It's already handling thousands of calls across dozens of MCP clients.

Thanks @rweiss57 @DustinPulver @LTzanani!

Read more in our blog: https://t.co/OMl1iBQvHo

[Tori Seidenstein @ToriSeidenstein]

Here’s the story of how we started Tadata—and what we’re building now at https://t.co/21YexWTGHq.

From an open-source project with 200K downloads to a full platform for MCPs, we’re helping people connect software to AI.

Built with @makhlevich, @tadata_it, and @ThatsShahar

[Tori Seidenstein @ToriSeidenstein]

Spinning up an MCP server for Claude Web Chat? Most backends don’t support SSE + token auth without OAuth.

Tadata does.

Here’s how I used it to auto-draft congrats messages for startup friends 👇

[Tori Seidenstein @ToriSeidenstein]

At Tadata, we’ve had a front-row seat as 2,000+ companies build MCPs—and it’s been full of surprises:
• 12% are 10K+ person enterprises
• Many aren’t “tech” companies
• Tons are building for internal use cases

Full blog post: https://t.co/NJtkukBwML

[Tori Seidenstein @ToriSeidenstein]

We’ve been heads-down at Tadata, and today we’re excited to launch something new 🚀

You can now spin up a hosted, authenticated MCP server with built-in analytics—starting from just an OpenAPI spec!

Here’s why that’s a big deal 🧵

[GitHub Projects Community @GithubProjects]

open-source tool to expose your FastAPI endpoints as Model Context Protocol tools with zero config.

Simple. Flexible. Production-ready.

[Miki Makhlevich @makhlevich]

@ThatsShahar just completely rebuilt FastAPI-MCP from scratch after the old architecture became a nightmare 😅 The big win? Precise control over MCP tool endpoints plus solid foundations for extension and composability—exactly what you asked for!

[Miki Makhlevich @makhlevich]

🚨New MCP servers alert!
The Model Context Protocol is changing how we work with #AI. We've gathered 4 new servers that you might not have heard of that dropped recently, demonstrating the diverse potential of this technology. 🧵 #MCP #AITools
1/6

[John Rush @johnrush]

What is MCP & why it's a big (huge) deal:

(model context protocol)

TLDR: MCP makes it possible for AI Tools to use external tools. E.g. Chatbot/IDE/AI-Agent can use Gmail/GoogleDrive/WeatherApp etc.

Detailed explanation for both, tech & non tech people (+demos):

1) AI Tools (chatbots, wrappers, agents, code generator, etc) wanna talk to external systems.

In pre-MCP world, one would have to write code to connect AI tool to the external system via API. Which meant every connection had to be pre-coded.

It also meant that every AI tool had to hard code its connection to every other tool. So if there are 1000 AI tools and 1000 external tools, then 1000000 hard-coded connections via API.

2) MCP is a standard protocol. This means that every AI tool has to implement this once, and then it can connect to thousands of external tools via this protocol.

3) The same goes for external tools. They all have to create an MCP server just once, and all AI tools that support MCP can connect to them.

4) It's a huge deal. Imagine 10k AI tools and 10k external tools now all have to implement MCP just once each. So it's 20k implementations. Versus 10k*10k=100M implementations.

5) This whole thing can also run on the cloud or on local computer.

See demos: