YASH SATYAWALI

At Voxomos.ai, I build and ship production GenAI and backend systems, including RAG chatbots, real-time voice bots, and LLM-based classification pipelines.

I've led a team of 4–5 interns and colleagues while improving latency and accuracy by up to 10x across RAG, voice, and inference pipelines through model and architecture optimization.

I work across LLM integration, semantic and hybrid search, vector infrastructure, model serving, and multilingual speech-to-text. My projects include document chatbots, local LLM and embedding-model benchmarking, transcription APIs across six-plus languages, and an LLM-powered phishing detection system.

I enjoy turning evaluated AI systems into scalable, high-performance applications for real-world use cases.