Open Source Hack Day: Surogate / Invergent AI

The Surogate AI Hackathon is the perfect opportunity to jump straight into the world of Large Language Models and work on real hardware-accelerated training, not just API wrappers. Surogate is built for developers and enterprises that need fast experimentation scalability and predictable outcomes whether running on-premise, in private clouds, or inside turnkey systems such as the DenseMAX Appliance. You’ll start with pre-built training recipes and have the chance to dive into Pre-training + Fine-tuning: full fine-tuning, LoRA/QLORA. It’s an ideal format both for those who want to learn how state-of-the-art AI infrastructure works leveraging Smart CPU Offloading for weights, gradients, activations, quants and for those who want to see what applied AI development looks like, with concrete impact and visible results during the hackathon itself. Beyond the low-level model optimization, the hackathon is also a launchpad for AI agent ideas that can turn your custom models into useful and creative applications. Empowering developer choices , once you lay the fine-tuning foundations using Supported models: Qwen2.5, Qwen3, Qwen3 MoE, Llama 3+, Nemotron Nano., you can start writing autonomous agents and contribute directly to the ecosystem alongside other participants passionate about LLMOps, machine learning, and AI tech. If you want to learn fast, work in a motivated team, and walk away with valuable hands-on experience in a real project, this hackathon is the right place for you. https://surogate.ai/ https://docs.surogate.ai/ https://github.com/invergent-ai/surogate