# Dipti Chaudhari

> Software Engineer (Search & ML Infra) | Semantic Retrieval • HNSW • LLM Embeddings • Distributed Systems

Location: Fremont, California, United States
Profile: https://flows.cv/dipti

I’m a Software Engineer specializing in Search & ML Infrastructure with 7 years of experience building low-latency, high-throughput distributed systems. My work sits at the intersection of backend systems and machine learning, with a focus on semantic search, ranking, and production ML serving.
Recently, I’ve been leading query-side semantic retrieval using LLM embeddings + HNSW, building end-to-end pipelines from embedding generation and storage to ANN search and deep learning ranking, and deploying models via ONNX + NVIDIA Triton to meet strict p99 latency and throughput SLAs.
A few examples of impact:
- Designed and deployed semantic retrieval for search using LLM embeddings + HNSW, enabling real-time semantic search at scale with ~82% latency reduction (p99 < 20 ms) via caching and optimized serving.
- Built and productionized deep learning ranking models for search/navigation, improving relevance and downstream shopping metrics on a high-traffic retail portal.
- Standardized ONNX model serialization + Triton-based serving, creating a reusable inference stack that shortened model iteration cycles and simplified cross-team adoption.
- Worked across teams to improve operational reliability (on-call, dashboards, incident response), reducing error rates and improving availability for critical search flows.

## Work Experience
### SDE II- Search Engine Technologies @ A9.com
Jan 2022 – Present | San Francisco Bay Area
I work on the search & ML infrastructure that powers a large-scale e-commerce search experience, focusing on semantic retrieval, ranking, and low-latency serving.

### SDE- Search Engine Technologies @ A9.com
Jan 2019 – Present | San Francisco Bay Area

### Data Scientist- Visual Search @ A9.com
Jan 2018 – Present | San Francisco Bay Area
• Optimized data annotation tasks for data used for Amazon Visual Search pipeline for Amazon Prime mobile app.
 • Built Machine Learning models for retrieval system, recommendation system and attribute prediction tasks.
 • Python-Tensorflow, Sklearn, AWS-DynamoDB, EC2, ElasticSearch, Lambda

### Machine Learning Research Assistant @ Human Performance and Robotic Laboratory
Jan 2017 – Jan 2017 | Greater Los Angeles Area
• Led research on Deep Learning in rehabilitation robotics of upper limb prosthetic.
• Academic paper published in ISER 2018. Presented in Robotics: Science and Systems 2017 at MIT
• Speaker at “ 40-years of Operational Space Symposium ” at Stanford University 2017
•  Python, Matlab, Tensorflow, Matplotlib, C++, Kinova Jaco robot arm SDK

### Machine Learning Intern @ Nonpariel Capital
Jan 2017 – Jan 2017 | Fremont, CA
• Implemented Stock price time series prediction application for using Recurrent Neural Networks.
• Python, Tensorflow, Keras, Matplotlib

### Software Engineer @ Great Software Technology
Jan 2013 – Jan 2015 | Pune Area, India
• Led automation team for VOIP product on web, server and client side.
• Reduced 80% time and 100% manpower by automating the test plans for US based Internet Telephonic company.
• Python, Perl, HTML, shell, Robotframework, Appium

### Software Developer @ InTouchApp
Jan 2012 – Jan 2013 | Pune Area, India
•  Improved the batch processing remote object monitoring system by real-time object monitoring system via Android app.
• Published the research paper for the work in IJMECE 2013.
• Python, HTML, CSS, AJAX, Java, Android SDK


## Education
### Master’s Degree in Computer Science
California State University, Long Beach

### Bachelor’s Degree in Computer Science
Pune Institute of Computer Technology


## Contact & Social
- LinkedIn: https://linkedin.com/in/diptikchaudhari
- Portfolio: https://sites.google.com/view/diptikchaudhari/
- GitHub: https://github.com/diptichaudhari24

---
Source: https://flows.cv/dipti
JSON Resume: https://flows.cv/dipti/resume.json
Last updated: 2026-04-10