Open to opportunities

Ayush Tiwari

@ayushtiwari9

Message

I build scalable full-stack, AI, and backend systems that automate complex workflows.

India

Message

What I'm looking for

I'm looking to build scalable backend, full-stack, and AI-powered products where I can automate complex workflows, improve system performance, and grow through hands-on engineering challenges.

I've built production-focused AI and backend systems at Genpact, Samsung PRISM Research, and SRM Technologies. At Genpact, I architected a Jira webhook-driven autonomous API QA pipeline using Python and LangGraph that reduced repetitive manual API testing effort by 65%.

I also fine-tuned Llama-2-7b for sensor anomaly detection at Samsung PRISM Research, improving accuracy by 12% over baseline and reducing inference latency and compute cost by 35%. Across freelance and internship work, I've delivered multi-platform marketplace applications, AI video-conferencing systems, REST services, and scalable Java/Spring Boot projects using React, Node.js, PostgreSQL, Redis, Docker, and Kubernetes.

Experience

Work history, roles, and key accomplishments

Software Engineer Intern

Genpact India Pvt. Ltd.

Jan 2026 - Jun 2026 (5 months)

Architected a Jira webhook-driven autonomous API QA pipeline using Python and LangGraph, reducing manual testing effort by 65%. Built a RAG-based retrieval layer and rule-based validation engine to generate schema-aware test cases.

Python LangGraph PostgreSQL RAG API Testing Jira Webhooks Concurrent Programming

Freelance Software Developer

Self Employed

Jun 2025 - Dec 2025 (6 months)

Delivered a multi-platform grocery marketplace end-to-end, including mobile apps, backend, and admin dashboard, under an Rs.85K contract.

Ionic Capacitor Express MongoDB Android iOS Web Development

Research Intern

Samsung PRISM

Mar 2024 - Nov 2024 (8 months)

Fine-tuned Llama-2-7b for sensor anomaly detection, achieving 85% accuracy and reducing inference latency by 35% via quantization and CUDA optimizations.

PyTorch LLaMA 2 Fine Tuning Docker Kubernetes REST APIs CUDA Quantization

Software Engineer Intern

SRM Technologies

Feb 2024 - Aug 2024 (6 months)

Developed an AI video-conferencing backend with WebRTC optimizations, cutting latency by 45% and improving app load time by 30%. Automated meeting transcription and optimized SQL queries.

Express WebRTC Next.js SQL Performance Optimization