Rajan N

Applied AI/ML Engineer focused on productionizing LLM systems, including RAG pipelines and scalable backends. Skilled in evaluating model performance and failure modes, and optimizing reliability, cost, and quality trade-offs in production systems. Translates complex problems into structured AI workflows using AWS, FastAPI, and distributed processing systems.

Data Scientist (AI/LLM Engineering Focus) - Digtinctive India Pvt Ltd - Noida, UP

(2025-07)

• Led end-to-end development of an AI Insights platform, executing 12+ ML and LLM experiments and achieving 98% processing accuracy for text insight generation workflows.
• Built backend AI services using FastAPI, Docker, Redis, and Celery for high-throughput asynchronous inference.
• Benchmarked AWS Bedrock LLMs for RAG applications, comparing cost, accuracy, and response quality across Python generation, SQL reasoning, and document QA tasks.
• Identified retrieval bottlenecks and failure modes in RAG pipelines, improving response reliability and supporting model selection decisions.
• Proposed and implemented system-level optimizations that reduced compute cost by 30% and resolved critical edge-case failures in the processing pipeline.
• Led internal engineering knowledge-sharing sessions on Docker, Pytest, and AI engineering practices, improving team-level understanding of production ML workflows. Proposed an AI-first development workflow combining human validation and automated reasoning systems to improve engineering velocity and long-term productivity.

Bachelor of Technology - Information Technology - Global Institute of Engineering and Technology (GIET) (2021 - 2025)