Koushal Anitha Raja

Ai Software Engineer @Turing

US

Architected scalable Python backend using LangChain with RAG pipelines for LLM apps, delivering context-aware outputs in production environments.• Implemented RLHF and SFT to fine-tune LLMs for code generation with real-time loss, accuracy, and system metrics tracked for efficient optimization.• Built Gemini API wrappers using Node.js and MongoDB for fine-tuning workflows and low-latency access to LLMs under concurrent load conditions.• Designed evaluation pipelines with Hugging Face evaluate and custom metrics to benchmark model quality against baselines and annotated datasets.• Deployed reproducible LLM CI/CD through Docker, Kubernetes, and GitHub Actions, ensuring rollback support and zero-downtime model releases.• Orchestrated modular data pipelines in Apache Airflow to clean, structure, and track datasets (JSONL, CSV, code), streamlining model training input.

R.M.D ENGINEERING COLLEGE

Bachelor of Engineering - BE, Information Technology

Stevens Institute of Technology

Master of Science - MS, Computer Science