Headline

Senior Data Scientist | Generative AI, LLMs, Agentic AI, Computer Vision, Demand Forecasting, Pricing Optimization | Python, Azure, Databricks | Building Production AI Systems

About

I'm a Senior Data Scientist with 10+ years of experience building AI and Machine Learning solutions that solve complex business problems across SaaS, Retail, CPG, and Oil & Gas industries. My work spans the full AI lifecycle—from problem definition and data engineering to model development, deployment, and business adoption. Over the years, I've delivered solutions in Demand Forecasting, Pricing & Promotion Optimization, Computer Vision, NLP, and Generative AI, helping organizations make data-driven decisions and automate critical business processes. I enjoy translating complex data into practical, scalable solutions that create measurable business impact. My core expertise includes: Generative AI: Large Language Models (LLMs), Retrieval-Augmented Generation (RAG), Agentic AI, Prompt Engineering, MCP Machine Learning & Deep Learning: Predictive Modeling, Time Series Forecasting, NLP, Computer Vision, Deep Learning Production AI Systems: Model Deployment, MLOps, Real-time Inference, AI Solution Architecture Technical Stack: Python, SQL, R, Databricks, Azure Machine Learning, MongoDB, NVIDIA DeepStream Throughout my career, I've worked on a wide range of AI initiatives—from forecasting demand for global retail brands and optimizing pricing strategies to developing real-time computer vision systems for intelligent transportation and industrial surveillance. More recently, my focus has shifted toward building production-ready Generative AI applications, exploring AI Agents, multi-agent systems, RAG architectures, and enterprise AI solutions. With a strong academic foundation in Mathematics and a passion for continuous learning, I enjoy solving challenging problems where data, engineering, and business intersect. I believe the most impactful AI solutions are not only technically sound but also practical, scalable, and aligned with business goals. I'm always interested in connecting with professionals working on AI, Machine Learning, Generative AI, and Data Science, and I'm open to collaborating on innovative projects that push the boundaries of applied AI.

Location

Pune District

AI stack

antigravity

claudecode

Work experience

Data Scientist at Siteminder, Jun 2026 - Jul 2026 · 2 mos, Pune District

Senior Data Scientist at Fractal, Apr 2024 - May 2026 · 2 yrs 2 mos, Pune District

Senior Data Scientist at Nihilent, Nov 2022 - Apr 2024 · 1 yr 6 mos

Data Scientist at Reliance Industries Limited, May 2020 - Nov 2022 · 2 yrs 7 mos, Mumbai, Maharashtra, India, Usecase1: Wrong Direction detection of vehicle Project Description: Develop a real-time intelligent transportation system (ITS) to detect vehicles traveling the wrong way on the road. The concept of this wrong-way system is to detect such vehicles as soon as they enter an area covered by a single closed-circuit television (CCTV) camera. The developed system is based on video imaging and covers three aspects: detection, tracking, and validation. Algorithms Used: YOLOV3. Usecase2: Count of footfall at different sites/locations. Project Description: Build a robust system which will count whether people were coming into or going out as well as the time they entered or left in JWD/JWC and for Jamnagar manufacturing division. Algorithms Used: NVIDIA People Net. Model had an accuracy of around 90-95%. Usecase3: Identify the wrongly parked vehicle in an unauthorized zone. Project Description: Illegal parking is a critical problem everywhere. For a better traffic management, identify the vehicles which are parked in non-parking zones for that build a solution which can not only identify the wrongly parked vehicles but also raise an alarm so that security people can be alerted or due to the raised alarm concerned car can be alerted. Algorithms Used: NVIDIA TrafficCamNet (Primary Engine) and Vehicle TypeNet(Secondary Engine). Model had an accuracy of around 75-85%.

Senior Data Science Analyst at Accenture, Jun 2018 - May 2020 · 2 yrs, Mumbai Area, India, Project-2: Problem Statement : Demand Planning for Diageo India across all categories. Responsible for below activities, 1. Luxury category products ordering from Scotland. 2. Analysis of Root to Market of products. 3. Analysis of current trends in the market with Demand Planner Tools Used - R, Excel Algorithms Used - ARIMAX, NNAR, XGBOOST, Random Forest. Project-1: Problem Statement : Demand Planning for e-Comm applications for Ascena Retail Group(USA). Responsible for end to end activity i.e. 1. Data extraction 2. Data cleansing 3. ADS creation (Analytical Data Set) 4. Model building,tweaking,and result interpretation. 5. Generation of Final results Tools Used - Base SAS, R, Python, Excel. Algorithms Used - ARIMAX, NNAR, XGBOOST,LSTM

Education

Master of Science (M.Sc.), Mathematics, National Institute of Technology Rourkela