Headline

Research Engineer (Founding Team) | Foundation Models, Voice AI & Agentic Systems | LLM Fine-Tuning • Synthetic Data • Evaluation & Observability

About

I build and deploy production-grade LLM and NLP systems with a strong focus on scalability, efficiency, and real-world reliability. At Gnani AI, I’ve worked on multilingual LLMs and conversational AI systems serving millions of users, taking models from research and fine-tuning through optimized inference and monitoring in production. My work includes fine-tuning and optimizing 4B–27B parameter models using PEFT and quantization, deploying high-throughput inference with vLLM, and building robust ML testing and evaluation frameworks (A/B testing, regression testing, MLflow). These efforts resulted in ~48% cost reduction and ~20% inference speed improvements in production systems. I enjoy operating at the intersection of LLM modeling and ML systems — improving latency, cost, and accuracy while making deliberate engineering trade-offs. My experience spans LLM fine-tuning, RAG and search systems, multimodal models, and production MLOps, with a strong bias toward building systems that scale and ship. Core focus areas: LLM & NLP systems · Fine-tuning and model optimization · RAG and search · Scalable inference · ML systems engineering

Location

India

AI stack

anthropic apis

claude

deepseek

elevenlabs

hermes agent

hugging face

Work experience

Founding Research Engineer at SuperBryn, Jan 2026 - Present · 8 mos, Bengaluru

NLP Engineer at gnani.ai, Jun 2024 - Dec 2025 · 1 yr 7 mos, India

Artificial Intelligence Engineer at Meroku, Mar 2024 - Jul 2024 · 5 mos, India

Content Writer at Entwicklers, May 2021 - Jan 2023 · 1 yr 9 mos

Education

Bachelor of Technology - BTech, Computer Science, J.K. Institute of Applied Physics and Technology

Public links

github: https://github.com/imcoza