Looking for a job

Kshitij Sharma

@kshitijsharma8

Lead AI/ML Engineer, Data Scientist & AI Architect with 12+ years in production AI: computer vision, GenAI, recommendations & edge AI.

India

Message

What I'm looking for

I’m interested in senior technical AI/ML roles where I can combine hands-on ML/data science, strong software engineering, systems architecture, and end-to-end technical ownership.

I’m a Lead AI/ML Engineer and Data Scientist with 12+ years of experience across software engineering, data science, machine learning, and production AI.

My work spans computer vision, predictive and probabilistic modeling, recommendation and search systems, GenAI/RAG, and edge AI. I work hands-on across the full lifecycle—from problem formulation, data and model development to production software and system architecture.

At Mercedes-Benz R&D North America, I developed personalization and predictive ML systems and took models from Python research through production C++ deployment in vehicles. At Wynk, I built recommendation and search systems serving 50M+ users. More recently, I’ve been building an end-to-end computer vision and retail analytics platform spanning real-time edge processing, cloud services, and predictive analytics.

I started my career in software engineering and hold an M.S. in Computer Science from NC State, with published research at the SIAM International Conference on Data Mining.

I’m interested in Lead, Staff, Principal, and AI Architect roles where I can combine hands-on ML/data science, strong software engineering, architecture, and technical leadership.

Experience

Work history, roles, and key accomplishments

Current

Lead AI/ML Engineer & Architect

Current

Independent AI Product / Computer Vision Retail Analytics Pl

Jul 2025 - Present (1 year 3 months)

Architected and built an independent multi-camera computer vision platform for live retail environments, integrating vision-derived customer analytics with POS and operational data. Engineered a multi-camera pipeline using YOLO12s, Kalman/Hungarian tracking, and OSNet/ArcFace with FAISS indexing, sustaining 65-85 ms median per-camera latency on 2.7K feeds.

Computer Vision Edge AI YOLO Object Detection Trackings Faiss Django fastAPI PostgreSQL Docker CI CD Cloudflare REST APIs RAG Architecture

Lead Data Scientist

Mercedes-Benz R&D North America

Apr 2021 - Jun 2025 (4 years 2 months)

Led development of embedded AI and personalization systems for Mercedes-Benz vehicles, applying Bayesian inference, variational modeling, causal inference, and ML to driver behavior, comfort, climate, and EV use cases. Owned the full lifecycle from Python research to production C++ implementation, with software deployed directly on 2026 Mercedes-Benz production vehicle models.

Machine Learning Bayesian Statistics Clustering predictive modeling Python Embedded Systems C++ Programming Docker Azure Data Science

Lead Data Scientist

Wynk Limited

Apr 2020 - Mar 2021 (11 months)

Built large-scale recommendation and search systems for Wynk, serving 50M+ users. Developed hyper-personalized playlist models, music embeddings, and learning-to-rank search using deep NLP and Elasticsearch. Engineered TensorFlow/Spark pipelines combining audio features (MFCC) and user behavior to power song similarity, personalization, and ranking.

Python Spark Kafka Recommendation Systems Learning To Rank Search Deep Learning Embeddings Transformers

Machine Learning Engineer

Mercedes-Benz R&D North America

Aug 2016 - May 2019 (2 years 9 months)

Built embedded ML and predictive systems for Mercedes-Benz vehicles using Bayesian inference, nonparametric methods, and mixture models. Developed location clustering, automation, and in-cabin passenger analysis, then implemented and tested models in production C++ for real-time deployment on vehicle head units.

Python C++Machine Learning Data Science Clustering Bayesian Statistics Causal Inference Probability Calibration predictive modeling

Software Engineer, Research Intern

SAS Institute Inc.

Jun 2015 - Apr 2016 (10 months)

Developed backend features for SAS’s enterprise analytics and data visualization platform within a large-scale C++ graph framework. Enhanced Graph Template Language, implemented new SAS procedures and visualization capabilities, and migrated graphing functionality from Java to the newer C++ platform.

C++SAS Data Visualization Analytics

Software Engineer

Publicis Sapient

Nov 2011 - Jun 2014 (2 years 7 months)

Built backend services for a large-scale multi-channel e-commerce platform in Java, with ownership of order capture, payments, and shopping-basket workflows. Designed web service integrations for end-to-end commerce transactions and developed an in-memory inventory caching layer to improve transaction speed and reduce repeated backend/database lookups.

Java Web Services RDBMS Software Development Ecommerce