About
AI/ML Full Stack Developer with 4 years of hands-on experience in GenAI, large language models (LLMs), and backend microservices architecture. Demonstrated expertise in developing scalable solutions using Python, Flask API, LangChain, and Retrieval-Augmented Generation (RAG). Known for implementing the factory design pattern and Airflow for optimized data processing and API efficiency. Proven track record of enhancing model performance, boosting data accuracy, and strengthening client security through a metadata-driven approach.
Experience2
Senior AI engineer
Jan 2025 — PresentNTT Data
Designed and implemented deployment architecture for iAgentOps platform on Azure Kubernetes Service enabling scalable GenAI agent deployment. Developed AI-powered Interview Tool to improve employee client-readiness. Contributed to iAgentOps observability platform integrating Loki, Jaeger, Prometheus, and Grafana. Mentored trainees in delivering multiple GenAI proof-of-concepts.
AI/ML Full Stack Developer
Jan 2022 — Dec 2025Accenture
Led backend team to architect on-the-fly visualization with Agentic workflow, reducing data analysis costs by 10%. Engineered AI chatbot using LangChain and RAG, reducing SQL generation time by 40%. Built and optimized DevOps pipelines for GCP and Azure, boosting productivity by 15%. Used metadata-driven responsible AI architecture, improving client data protection by 80%. Created helper functions with Factory Pattern, boosting data quality by 50% and cutting API costs by 30%. Developed micro-service with LangChain and RAG to automate security questionnaires, saving 80% client time.
Projects2
Drowsiness Detection System
Formulated a real-time drowsiness detection system with CNN and OpenCV, achieving 95% accuracy in classifying drowsy and non-drowsy states. Implemented a robust alert mechanism with visual and auditory notifications, including low-frequency sound, resulting in a 10-15% increase in user awareness and responsiveness.
- Python
- CNN
- OpenCV
- Postgres
Facial Attendance System
Designed a facial recognition attendance system with OpenCV, achieving 70% improvement in real-time identification and logging accuracy. Established a data management solution using Postgres, improving retrieval speed of attendance records by 5-10%.
- Python
- Google Cloud Console
- OpenCV
- Postgres
Education2
Bachelor of Technology
2017 — 2021Medicaps University
Electronics and Communication
XII
2015 — 2016Gayatri Shiksha Niketan High School
Physics, Chemistry, Mathematics