Open to opportunities

Saumya Ranjan Sahoo

@saumyaranjansahoo

Message

Senior Python/LLM engineer building scalable backend and evaluation systems.

India

Message

What I'm looking for

I’m looking for a team where I can build production-grade Python/LLM backend services—design clean APIs, run rigorous evals with pytest, tune performance, and ship reliably in Agile with strong engineering standards.

I’m a Senior Python Developer and LLM Engineer with 5+ years of experience building production-grade backend systems, REST APIs, and AI-powered applications. At Turing, I’m currently engineering LLM training and evaluation tooling, combining reliability-focused testing with practical, high-impact delivery.

I’ve architected 5+ Django applications across e-commerce, fintech, and AI domains, optimizing database performance by up to 84% and improving API response times dramatically. I’ve built systems that serve 15K+ daily active users, delivered performance from 2.5s down to 400ms, and reduced production bugs by 60% through disciplined pytest coverage.

I’m energized by scalable architecture, microservices, and LLM pipeline quality—especially LangChain/RAG workflows, vector database integration, and prompt/evaluation design. I’ve mentored cross-functional teams of 6–9 engineers and led monolith-to-microservices migrations that increased deployment frequency by 3x while maintaining high client satisfaction.

Experience

Work history, roles, and key accomplishments

Current

LLM Python Engineer

Current

Turing

Sep 2024 - Present (1 year 9 months)

Architected Python/JavaScript workflows for LLM training and evaluation, improving model accuracy by 25% over 3 training cycles. Built OpenAPI-based RL Gym tooling and test suites with pytest (90%+ coverage), achieving 95% edge-case coverage and maintaining 98% client satisfaction.

Python fastAPI Pytest OpenAI API RL Gym OpenAPI Prompt Engineering SQL LLM Evaluation React

Senior Python Developer

Kogo Tech Labs Pvt. Ltd

Apr 2024 - Aug 2024 (4 months)

Led backend architecture and development for an AI-powered location assistant integrated with Mappls, serving 5K+ daily users with sub-500ms responses. Implemented FastAPI services with PostgreSQL and RAG (embeddings + vector DB), using Redis caching and query optimization to reduce API response times by 60% and database load by 40%; mentored junior developers to cut PR revisions by 45%.

Python fastAPI Django REST Framework PostgreSQL Redis RAG Vector Databases Embeddings GPT 4

Software Engineer (Python/Django)

Swagg Technologies Pvt. Ltd

May 2020 - Apr 2024 (3 years 11 months)

Developed and maintained 3+ production-grade Django applications and 40+ REST APIs across e-commerce, fintech, and real estate, serving 15K+ combined daily active users. Optimized 20+ PostgreSQL/MySQL queries to reduce average latency from 2.5s to 400ms (84%), built Scrapy-based scraping pipelines processing 10K+ records/day with 99.5% data accuracy, and reduced production bugs by 60% using pytest