Jerry Chi’s Post

For the Hashnode #AIForTomorrow hackathon, I launched https://congressgpt.app to help us finally understand what Congress is actually doing. Keeping up with what Congress is doing is hard. The language used in the bills can be challenging to follow. CongressGPT makes it easy to ask questions and understand congressional bills. It's more important than ever to stay informed about what our representatives are doing, and I hope CongressGPT helps you do just that.

At Lightning AI, we often get asked how I can take my pretrained or finetuned LLM, and successfully deploy it. This Studio below is a good example of deploying Meta's Llama-3 (8B) model behind a private API with LitServe, Lightning's new open source serving library. The Studio already comes preconfigured with the code, checkpoints and all required installed dependencies! It's batteries included. Built by creator of Pytorch Lightning himself, William Falcon. Check it out below. https://lnkd.in/eVJiReTn

I'm thrilled to introduce the Pass-in project, an API I created during the NLW Unite event by Rocketseat aimed at handling events and attendees. Pass-in makes it simple to set up events, enroll participants, conduct check-ins, and much more, all in an efficient and organized manner. By utilizing cutting-edge technologies such as Fastify, Prisma, and Zod, Pass-in provides a solid solution for streamlining event management. repo: https://lnkd.in/dmtDRuja #Passin #API #EventManagement #NLWUnite #nodejs #node

Check out the chat bot that I have been working on. I have used the following 1) langchain: a framework for developing LLM applications 1) langchain_huggingface : Interfacing hugging face endpoints and pipelines in langchain 2) Llama 3.2 1b instruct : Open source LLM from meta 3) gradio: a web interface for demonstration of the chatbot The application is hosted in huggingface_spaces. This is a work in progress. Have ideas to develop the chatbot into something unique. For collaborations, DM me https://lnkd.in/gfMhBZ7Y

It's performance review season. Are you ready? Sadly, most aren't... Good thing Adell & Ciprian (Chip) built PerfWriter, right? This project used agents to help solve a gnarly problem - building your case for a strong review season. With a roadmap that removes recency bias and can pinpoint that 2am on call fix you made, even if it was only 2 LOC. The duo really went hard to get this live and spent time grilling Saumil - I think I heard 'edges' and 'nodes' uttered at least 42 times. iykyk. At Squire AI hackathons it isn't just about technology. It's about the nitpicks you get to give Saumil along the way 🥰 Remember, by the end, can you guess who won: 1. First overall 2. Most technical 3. Made at least $1 4. Audience fav 2 Demos to go. I think Karl's taking side bets btw and odds are decent. He only takes something called BTC though... 🤔

Check out this demo of the Realtime API from our DevDay. My favorite part: voice is now available to experiment directly in the Playground. We also launched a less highlighted capability called "Generate Anything" which allows you to create detailed, optimized prompts, including functions and JSON, from basic inputs

Launching Mage.ai locally. The easy start ended up in a total disaster. I need to read more before making my first ETL... https://lnkd.in/dYui3KSy

Streaming intermediate events in RAG is crucial for best user experience 🚀 Let's see how to use LlamaIndex Instrumentation module, with server-sent events and Vercel AI SDK's StreamData feature to properly stream intermediate events to the frontend. Full tutorial under 3 minutes 🔥 Repo: https://lnkd.in/gK87Wmpc

"What do you do?" I will let panel Kseniia explain: In September 2021 I launched Elevatix. An AI engine that boosts revenue through personalized IAPs for each player. Making High Tech from AAA $Billion Companies accessible to all. Some recent client results: ✅ Improved repeat purchase rate by 22% 📈17% increase in IAP revenue. (FROM WEEK 2) 🐳 Increased ‘high-quality players' spending by 20% We also: 👥 Doubled average revenue per DAU 🧑💻 Maintain core team focus on new game developments 🏃♂️ Achieved significant revenue boost in days not months. All without: ⚙️ Distracting technical resources 🔧 Needing to Introduce new features 🤯 Changing the user acquisition strategy Not bad results for a 3-day integration. "You don't have to decide between high LTD and good retention rate" The integration is automatic, you just need to integrate our SDK. Everything including the AWS environment, we take care of. You don’t need any historical data either. ps. We’re currently running free trails 🎉 DM me and we can arrange a 1:1 Demo :)

Snowflake 🤝 Houseware 🤝 LangChain While building AI agents on Snowflake, we realized how the Langchain ecosystem didn't have established connectors to Cortex LLM models. Nipun & Shubham from the team recently built out a Chat Model integration for Snowflake Cortex inside Langchain's community libraries, and raised a pull request just earlier today(link in comments). This abstraction would allows developers to easily build AI applications on top of Cortex LLMs using the Langchain ecosystem, making Cortex a first-class citizen here onwards. OSS FTW 🚀