Parv Gatecha

Generative AI Engineer and M.Tech Computer Science candidate at IIIT Bangalore with hands-on experience building and deploying production-grade AI systems, LLM applications, RAG pipelines, multimodal AI systems, and agentic workflows. Experienced in Python, Hugging Face Transformers, PyTorch, FastAPI, Docker, Kubernetes, and cloud-native ML deployment. Built scalable AI applications involving fine-tuning, retrieval systems, evaluation pipelines, observability, and ML infrastructure.

Strong software engineering and systems design background with proven problem-solving ability as a Codeforces Expert.

NLP Research Intern at Augnito (2026-01 – Present)

• Built enterprise-scale LLM-powered Clinical Documentation Improvement (CDI) pipelines for healthcare workflows using transformer-based NLP systems on unstructured Electronic Health Record (EHR) data.
• Implemented Retrieval-Augmented Generation (RAG) architectures with semantic retrieval pipelines to improve factual grounding and reduce hallucinations in clinical AI outputs.
• Optimised medical Named Entity Recognition (NER) and text classification systems using domain-specific preprocessing and transformer fine-tuning techniques.
• Worked on production-oriented evaluation pipelines for AI quality monitoring, model reliability, and inference performance optimisation.

Applied Research Intern — Autonomous Driving AI at Bosch Global Software Technologies (2025-05 – 2025-07)

• Integrated production-grade Vision-Language Models (VLMs) into an autonomous driving perception pipeline, reducing inference latency by 15%.
• Evaluated and benchmarked 10+ multimodal foundation models across real-world driving scenarios using custom evaluation and observability frameworks.
• Designed scalable experimentation pipelines for multimodal AI systems involving scene understanding, object grounding, and instruction-following capabilities.

Software Engineer Intern at TatvaSoft (2024-01 – 2024-06)

• Built and deployed scalable backend services and REST APIs in.NET for a production appointment scheduling platform serving concurrent users.
• Optimised backend query performance and API response latency by 25% through efficient database design and caching strategies.
• Collaborated across engineering teams to ship production features with focus on maintainability, system reliability, and deployment readiness.

AI/ML Engineer — RLHF & LLM Evaluation at Scale AI (2022-01 – 2024-01)

• Contributed to RLHF, Supervised Fine-Tuning (SFT), and evaluation pipelines for large-scale conversational LLM systems deployed in production environments.
• Developed adversarial evaluation datasets and red-teaming workflows that reduced hallucination and reasoning failures by 30%.
• Worked on prompt engineering, model evaluation, alignment workflows, and data quality pipelines for transformer-based AI systems.
• Collaborated with cross-functional teams to improve model reliability, safety, and production deployment quality.

Master of Technology in Computer Science and Engineering – International Institute of Information Technology, Bangalore (2024-07 – 2026-06)

Bachelor of Technology in Computer Engineering – LJ Institute of Engineering and Technology (2020-07 – 2024-05)