Deploy your MCP Server to the cloud 🌥 . @digitalocean's App Platform makes it easy to deploy your MCP Server in 5 minutes. Check out this example of deploying a server that checks to see if domains are available, made by @amit 👇 .

3️⃣ Why standardization is crucial for building an interoperable AI ecosystem. If you're building AI tools or interested in the future of agentic AI, you should know about MCP! Check it out at the link below.

2️⃣How MCP shifts the burden of integration from developers to service providers.

The Model Context Protocol is revolutionizing how AI agents interact with external tools and services. In my latest article, I break down (with code examples):. 1️⃣The evolution from bare LLMs to tool-using AI agents.

📢 Just published: Model Context Protocol (MCP) - What it is, how it works, and why it matters 🧵.

Detect low confidence words in your Speech-to-Text to ensure you're feeding good inputs to your Agentic AI workflows.

NEW TUTORIAL: Build an AI agent with LiveKit for real-time Speech-to-Text 🤖. I've distilled everything you need to build an AI agent for real-time transcription into this video, using @livekit 's awesome Agents framework!.

RT @AssemblyAI: Build something cool with our Speech AI models and win $1,000!. We launched a challenge on @ThePracticalDev and we’re givin….

Build a free Whisper API with a GPU backend 👇. Check out my latest tutorial to see how you can leverage Colab's GPUs and Flask to build a free Whisper API that let's you run the largest models on GPU!.

In our tests, the next model update performs similarly to PhD students on challenging benchmark tasks in physics, chemistry, and biology. We also found that it excels in math and coding. In a qualifying exam for the International Mathematics Olympiad (IMO), GPT-4o correctly.

Florence-2 follows in the footsteps of LLMs, leveraging a unified architecture, simple training paradigm, and huge amounts of data to become competent at many different tasks. Learn more about Florence-2 and how to use it in my new blog 👇.

Florence-2, Microsoft's new Large Vision Model, is incredible! It can perform. 1. Captioning.2. Optical Character Recognition.3. Object Detection.4. Region Detection.5. Region Segmentation.6. Vocabulary Segmentation. (1/2)

Filter profanity from speech transcripts in just a few lines of code!.

🚀 How to build a Zoom app 🚀. Check out my latest video to learn how you can make a server-to-server Zoom app to automatically transcribe your meetings (uses OAuth).

Never wait for office hours again! Instead you can use AI to summarize and ask questions about your lectures 🤖. This app lets you ask questions about it with tailored answers that pull from the course material! Check out how to build it with Python 👇.

Don't forget - LLMs are inherently probabilistic and can subtly modify your inputs!. Here's some code I used to verify that timestamps I had provided to the model were not altered, check out the full tutorial in the video below:.

Did you know that you can automatically extract phone call insights in about 5 lines of code?. We accurately extracts a summary, action items, and contact information from a real phone call in this video using @AssemblyAI. Check it out!👇.

Automatically redact PII at scale with AI🔥🔥. Learn how to redact Personal Identifiable Information (PII) from audio and video files in less than 5 minutes using Python in this tutorial - the types of PII you redact are configurable!.

I wrote a Flask application for transcribing phone calls in real-time with @AssemblyAI @twilio and @ngrokHQ, feel free to check out the tutorial if interested 👇.

ICYMI - learn how Generative AI builds on principles from Physics.