At The Intellify, I build production conversational AI voice agents for a multi-tenant SaaS platform, combining streaming speech recognition, LLM tool calling, and text-to-speech over WebSockets. I’ve improved conversation quality with prompt engineering, guardrails, interruption handling, and real-time English-Spanish switching.
At Captus, I built computer vision pipelines for photo clustering, face recognition, and image quality assessment. I also trained a joint denoising and demosaicing U-Net and developed model evaluation and deployment workflows.
Earlier, as an AI Developer Intern at The Intellify, I built a RAG document question-answering system and developed a time-series forecasting module. My projects have also included vehicle detection and ANPR, graph neural networks, and data-driven imputation.
