Ankur Maddewad

AI/ML Engineer with hands-on experience building LLM, RAG, NLP, and deep learning applications. Experienced in Python, PyTorch, TensorFlow, Hugging Face, LangChain, FastAPI, vector databases, Docker, AWS, and CI/CD. Built and deployed AI applications involving semantic retrieval, LLM inference, model evaluation, prompt engineering, and ML pipelines.

Experience supporting NLP model fine-tuning through large-scale data annotation and validation.

AI Analyst — Contract at Innodata (2026-01 – 2026-06)

• Designed and optimized LLM prompts to improve reasoning accuracy and response quality across NLP and generative AI tasks.
• Performed large-scale data annotation, labeling, and validation for NLP/LLM model fine-tuning datasets.
• Contributed to instruction tuning, evaluation pipelines, and model alignment workflows for language-model development.
• Reviewed model outputs and applied quality-control guidelines to improve training-data consistency and model evaluation.

Technical Support Engineer at SCS Tech (2025-05 – 2025-10)

• Investigated application issues using logs, SQL queries, and monitoring tools, identifying root causes and resolving software defects.
• Troubleshot user-reported application issues and collaborated with cross-functional teams to deliver timely resolutions.
• Supported application reliability by analyzing recurring errors and escalating complex technical issues.

AI/ML Development Intern at TrustTech Innovation (2024-12 – 2025-05)

• Built a Retrieval-Augmented Generation (RAG) chatbot using transformer-based models and vector search for context-aware question answering.
• Developed FastAPI microservices for semantic retrieval and LLM inference.
• Containerized AI services using Docker and implemented CI/CD workflows for deployment.
• Integrated vector search and transformer-based inference into an end-to-end AI application.

B.Tech. in Artificial Intelligence and Machine Learning – G.H. Raisoni College of Engineering and Management, Pune (2025)