Open to opportunities

Pratyush Ojha

@pratyushojha

Message

I build scalable computer vision and LLM-powered AI systems.

India

Message

What I'm looking for

I seek hands-on roles building production ML/AI systems—particularly in computer vision and LLM integration—on collaborative teams focused on scalable deployment, model optimization, and measurable product impact.

I am a data scientist and AI engineer specializing in computer vision, NLP, and generative AI, with a strong focus on building scalable ML pipelines and full-stack AI applications.

I have developed real-time pose estimation pipelines using OpenCV and MediaPipe and engineered inference systems that leveraged Dynamic Time Warping to compare live poses to ground truth, achieving a 30% reduction in inference latency through model-loading and MediaPipe optimizations.

I built end-to-end projects such as an AI-powered interview simulation (MetaHire) and an AI fitness assistant (FITTAB), integrating Google Gemini and other LLMs for question generation and personalized plans, and deploying solutions with FastAPI, Express.js, MongoDB, Docker, Render, and Vercel.

I am a Kaggle Expert with a strong record of public notebooks and competition work, and I bring practical experience in model optimization, prompt engineering, and secure inference workflows to teams seeking impactful, production-ready AI solutions.

Experience

Work history, roles, and key accomplishments

AI/ML Intern

Aadiyog Healthtech Pvt Ltd

Jan 2025 - Apr 2025 (3 months)

Developed pose estimation pipelines using OpenCV, MediaPipe, and FastAPI for real-time yoga assistance and engineered a DTW-based inference system to compare live poses against ground truth. Optimized model loading and MediaPipe usage to reduce inference latency by 30%.