I build AI systems that actually work in production, not just in demos. RAG pipelines, multi-agent orchestration, memory architectures, real-time voice inference — end to end, from research to deployment.
I'm an AI Engineer specializing in building scalable, production-grade intelligent systems. My work sits at the intersection of LLMs, backend engineering, DevOps, and data science, translating research into reliable systems that reduce friction for real people, not just automating tasks, but anticipating needs.
Currently at Mira Copilot, I architect end-to-end pipelines spanning retrieval-augmented generation, multi-agent orchestration, dual-memory architectures, and real-time voice interfaces. I also own the deployment side, cloud infrastructure, containerization, and CI/CD from dev to production.
Before focusing on AI, I sharpened my analytical foundation through data consulting at FedEx Ground (via Katz / KPMG, placing 2nd in the KPMG US Cup) and academic research at the University of Pittsburgh, where I graduated with a 3.9 GPA in Information Science.
Full-stack AI travel rewards optimizer that helps users decide when to pay with points vs. cash. Built a deterministic decision engine evaluating loyalty program reward valuations to generate optimal booking recommendations, with FastAPI backend and Next.js frontend.
Core backend for a voice-first AI scheduling assistant. Architected end-to-end RAG pipeline, dual-memory system, ReAct agent orchestration with intent classification and tool routing, real-time STT/TTS streaming, and a calendar intelligence engine with explainable meeting suggestions.
A retrieval-augmented generation bot for Discord that answers questions grounded in custom knowledge sources. Built with Python, integrating vector search and LLM reasoning for accurate, context-aware responses inside Discord servers.
AI-powered desktop assistant built with Python and OpenAI's GPT-4, designed to engage in human-like conversations and perform automated tasks. An early exploration into conversational AI and task automation on local machines.
Full-stack photo gallery platform with user authentication, role-based access control, photo upload and metadata management, commenting, and admin moderation. Focused on responsive, device-friendly UI design.
Data analytics consulting project optimizing last-mile delivery for FedEx Ground. Built real-time Tableau dashboards, modeled 15% efficiency improvements, identified 9.8% projected cost savings through volume-density analysis. 2nd place, KPMG US Cup Competition.
Mobile app for real-time sign language to English conversion, built at Duke University's AI research lab. Applied computer vision and gesture recognition to break communication barriers for the hearing-impaired community.
Led a 4-member team enhancing the Hunger Game Search meta-heuristic algorithm — improving computational speed and mitigating premature convergence. Assessed with k-means clustering. Published in Webology, September 2022.
GPA: 3.9 · Specialization in AI, data science, and health informatics. Worked as Applied Data Scientist (Research) and Course Coordinator for the Master's in Data Science program.
Published research in optimization algorithms. Led award-winning project team — Best Project at Pradarshana for VR-based physical rehabilitation for cerebral palsy patients.
I'm open to AI/ML Engineer and Software Engineer roles where I can build things that matter — systems that reduce friction for real people.
Whether you have a role in mind, a collaboration idea, or just want to connect — I'd love to hear from you.