Sujit Nalawade — Senior ML Engineer
Open to Senior Roles
Sujit
Nalawade
Senior Machine Learning Engineer specializing in Computer Vision, Deep Learning, and scalable AI inference systems — building things that see and understand the world.
Sujit Nalawade
Senior ML Engineer @ Adagrad AI
📍 Pune, India
🎓 PGD Data Science — IIIT Bangalore
🎤 NVIDIA GTC Speaker (2021, 2022)
📧 technosujit25@mail.com
📱 +91 8329132283
// About Me
Building AI that
sees the world
I'm a Senior ML Engineer with 7+ years of experience building and deploying machine learning systems at scale. My work sits at the intersection of computer vision, deep learning, and production engineering.
From real-time video pipelines processing thousands of camera feeds to inference servers handling high-throughput batched requests — I specialize in taking research-grade models and turning them into robust, production systems.
I've spoken at NVIDIA GTC twice, sharing work on object detection at 1770 FPS and geometric deep learning with TensorRT.
// Work History
Experience
- Design and develop scalable Computer Vision and NLP pipelines for real-world production deployments.
- Lead cross-functional teams designing solutions handling 1000+ live camera feeds globally.
- Architect high-performance inference systems using NVIDIA Triton, TensorRT, and DeepStream SDK.
- Trained Word2Vec and Gen-sim embeddings for specialized insurance document corpora.
- Improved NER F1 score from 0.60 → 0.85 using BiLSTM + BERT architectures.
- Built a Doc2Vec-based document classifier achieving F1 score of 0.93.
- Developed a data masking library for anonymizing PII in XML documents at scale.
- Designed a payload management web interface using Angular and Node.js.
- Certified: Microsoft Azure — "Developing Microsoft Azure Solutions".
// Selected Work
Projects
🎥
DNSTAR
High-performance deep learning inference pipeline for real-time video processing. Deployed globally, monitoring 1000+ cameras with state-of-the-art object detection and classification.
👤
REID & MATCHING
Cross-camera person re-identification by matching feature vectors. Reduced execution time from 72 hours → 3 hours via multiprocessing and multithreading optimizations.
🛒
CHECKOUT ACCURACY
Video pipeline detecting unbilled goods at checkout. Deployed across 200+ registers in US and Canada, significantly reducing shrinkage and improving inventory accuracy.
🚗
GATEGUARD
Real-time vehicle number plate identification deployed across multiple sites in India, enhancing security through accurate vehicle logging in varied conditions.
🏠
NOBI ENGINE
Oriented bounding box detection pipeline monitoring elderly residents in care homes. Detects falls and unusual activity, alerting caregivers in real time.
⚡
TRITON INFER ENGINE
Inference API server supporting multiple models on NVIDIA Triton with batched inference, optimizing GPU utilization for high-throughput production AI workloads.
💬
TAXBOT
RASA-based chatbot for tax-related queries deployed on AWS EC2, combining NLU model training with conversational AI for accurate, user-friendly assistance.
🏪
POSEIDN
Pose estimation pipeline for retail analytics — estimates customer interactions and shopping behavior to help optimize store layouts and improve experience.
// Technical Stack
Skills
Expertise
Computer VisionObject DetectionImage ClassificationNLPDeep Learning
ML / Vision
PyTorchPyTorch LightningCUDATensorRTONNXRuntimeScikit-LearnOpenCV
NVIDIA Stack
DeepStreamTritonGStreamer
NLP / LLM
HuggingFaceSentence TransformersRASA
Backend / DevOps
Django RESTFlaskDockerGitHub ActionsPostgreSQLMySQL
// Background
Education
Post Graduate Diploma in Data Science
International Institute of Information Technology, Bangalore
Sep 2018 – Sep 2019
Bachelor of Computer Engineering
University of Pune, Pune, India
Jun 2012 – Jul 2016
// Public Speaking
Conference Talks
GTC 2022
Save Thousands of Man-hours with Geometric Deep Learning and TensorRT
NVIDIA GPU Technology Conference
GTC 2021
Object Detection at 1770 FPS: Designing CNNs that Scale Using DeepStream SDK
NVIDIA GPU Technology Conference