Addy AI

🎊 Thrilled to share that Joao Aires de Matos has joined us at Addy AI as our Founding AI Engineer. Jay was a Senior Software Engineer and tech lead at Deloitte and has a ton of experience building enterprise tech. He holds a Master's degree in Computer Engineering from the University of Aveiro and fun fact -he taught an AI course this past semester. I’ve been asking him to join us for a while, but as luck will have it – now’s the right time! Jay's an absolute rockstar, and I'm super excited to have him on!

🤩 I’m super excited to announce that Spatika Ganesh has joined Addy AI as our Founding Machine Learning Engineer. Spatika previously worked at Fidelity Investments on machine learning and has a ton of experience building products in the financial services industry. Spatika also graduated from Carnegie Mellon University with a Master’s in Electrical and Computer Engineering focusing on AI. While at CMU, she was a teaching assistant for the Deep Learning course, where she not only worked on neural networks but helped other students as well. Welcome to the team, Spatika! I can’t wait to see the awesome AI features you add to Addy AI ✨

Exciting news! 🌟 Today, we're launching the Docs for LangDrive. https://docs.langdrive.ai LangDrive is an open-source library to train 100+ Open LLMs with just a single interface. It handles everything from ingestion to training and deployment right from your terminal, calling our API, or as a package in your codebase. 🎯 Docs highlights 1️⃣ Get started Quickly get started with LangDrive and train your first model in a few minutes using the examples provided. 2️⃣ Data Connectors Data connects are in-built utils to connect your private data. From CSVs to JSON objects, Firebase Firestore, Emails, Google Drive, etc... All built-in to LangDrive. 3️⃣ Models Grab a base model from Hugging Face or select from our list of supported models. LangDrive provides you with what you need to run the training. From environment configuration to Docker images, etc. 4️⃣ Deploying Your Model Deploy your trained model directly to Hugging Face using LangDrive. Very soon you will be able to deploy to other services as well. 👥 Contributing Come build the future of AI with us. LangDrive is open source, and we welcome contributions from the community. There's a bunch of fun, exciting, and technically challenging work to be done. If you want to accelerate the accessibility of LLMs, join us! Make a PR at https://lnkd.in/g2wb_7jP or DM me for a quick onboarding. Charles Karpati and I are super excited about LangDrive and we think this is one of the most important pieces of work needed in the world right now. https://docs.langdrive.ai #LangDrive #AI #LLMs

⚡LangDrive update: Fit model footprint in CPU if GPU memory is unavailable or insufficient. 🛠️Train 100+ LLMs on your private data directly from your terminal. 🔥Many 7B or fewer parameter models can be finetuned on a CPU. but most of these models are defaulted to work with GPUs. In the latest LangDrive update, we’ve made it possible for smaller models like Falcon 7B to adapt to a CPU if GPU memory is unavailable. This means you can finetune it on a regular 8-core 32 GB RAM CPU 🤯 locally or in your cloud. ↪️In the end you can choose to get your weights or an endpoint to interact with the model. Check out the update: https://lnkd.in/g7JG2Dfw

📢 AI Developers, after the updates from OpenAI yesterday, it's very evident that the real value of AI lies in the opportunity to train LLMs on proprietary data. Charles Karpati and I have built LangDrive to help you train 100+ LLMs on proprietary data built-in with everything you need from ingestion to data connectors, privacy and compliance, and cross-border data protection. We initially built LangDrive as an internal tool at Addy AI but have decided to make it open-source. Start using LangDrive today for your projects at https://lnkd.in/g_9eXVej Please DM us if you need any support getting started with using LangDrive in your projects.

Last Friday, SVLG held its inaugural Retreat: Ideas, Innovation & Priorities presented by Amazon and hosted by Synopsys. We had the pleasure of hearing from industry experts, thought leaders, and decision-makers focusing on technology and innovation both in Silicon Valley and around the world. A special thank you to our speakers from Addy AI, Amazon, Bank of America, Blue Shield of California, IBM, and Synopsys Inc. The morning event was engaging and provided insights on current cutting-edge innovations that are driving advancements in the innovation economy. #SVLGRetreat