- 🔭 I'm interested in LLM and AudioLLM
- 🌱 I'm currently working at Continuum AI, where we recently released our in-house end-to-end model FlashLabs Chroma 1.0
- 🎧 My work spans large-scale data pipelines, GPU training, and real-time audio / TTS systems
- 💬 Feel free to reach out about multimodal speech models and audio infrastructure




