Open to opportunities

Priyank Tyagi

@priyanktyagi

Message

I build production full-stack web apps, AI features, and scalable APIs with measurable performance gains.

India

Message

What I'm looking for

I'm looking to build production web products where I can own work across frontend, backend, cloud, and practical AI features while growing with a strong engineering team.

I've built production SaaS and AI-enabled web applications end to end, from React and Next.js interfaces to FastAPI, Node.js, and PostgreSQL services.

At AgroLogi Technologies, I built a multi-tenant Next.js application with RBAC, i18n, and support for three markets. I reduced API response times by 40%, lowered ML detection latency by 35%, improved accuracy by 25%, and cut deployment time by 60% through Docker and GitHub Actions CI/CD.

As a Research Intern at IIT Roorkee, I replaced static validator selection in a blockchain consensus protocol with a gradient-boosted ML model, improving throughput by 30% and reducing block-confirmation latency by 20% on a 20-node testnet.

I've also built document-QA, real-time customer-support, and component-generation products, filed two patents, and earned 10+ hackathon wins. I enjoy shipping practical systems where AI, clean architecture, and performance improvements create clear user value.

Experience

Work history, roles, and key accomplishments

Full Stack Developer

AgroLogi Technologies

Sep 2025 - Feb 2026 (5 months)

Built a multi-tenant Next.js application with RBAC and i18n, cutting API response times by 40% via Redis caching. Designed FastAPI backend services including an ML inference endpoint, containerized with Docker, and automated CI/CD with GitHub Actions.

Next.js React RBAC Redis fastAPI Machine Learning Docker GitHub Actions CI CD Vercel Render Faiss GPT 4

Research Intern

Indian Institute of Technology (IIT) Roorkee

Apr 2025 - Aug 2025 (4 months)

Replaced static validator selection in a blockchain consensus protocol with a gradient-boosted ML model, improving throughput by 30% and reducing block-confirmation latency by 20% on a 20-node testnet.

Blockchain Machine Learning Gradient Boosting