Pankaj Agarwal

Pankaj Agarwal

Senior Software Engineer - Machine Learning

Uber

Pankaj Agarwal is a seasoned Machine Learning Engineer with nearly 12 years of experience designing and deploying data-driven solutions at scale. Currently at Uber, he focuses on building advanced search and recommendation systems for UberEats, tackling complex problems in personalization and information retrieval.

Pankaj’s previous roles at Compass, Myntra (a Flipkart Group company), and FICO have centered around developing robust machine learning pipelines and predictive models across e-commerce and financial domains. He has also published research at top-tier conferences such as KDD, contributing to the academic and applied ML community alike.

Pankaj holds a Bachelor of Technology from the Indian Institute of Technology, Kharagpur, and a Master’s in Computer Science from Georgia Tech. His expertise spans machine learning, deep learning, and statistical analysis, with hands-on skills across Python, SQL, Hive, MongoDB, and modern cloud platforms.

In the world of online food delivery, user search queries are often vague, incomplete, or noisy — like "best pizza", "veg thali under 200", or "birynai" (yes, with a typo). This talk explores how Retrieval-Augmented Generation (RAG) can help rewrite such queries into more precise and intent-aware forms, improving both relevance and user experience. We’ll cover the core concepts behind RAG, how it combines external retrieval with generative language models, and how it compares to traditional query rewriting approaches. The session will wrap up with a hands-on demo showcasing a real-world use case in the online food delivery space, illustrating how RAG can be used to bridge the gap between user intent and search results.

Read More

Managing and scaling ML workloads have never been a bigger challenge in the past. Data scientists are looking for collaboration, building, training, and re-iterating thousands of AI experiments. On the flip side ML engineers are looking for distributed training, artifact management, and automated deployment for high performance

Read More

Managing and scaling ML workloads have never been a bigger challenge in the past. Data scientists are looking for collaboration, building, training, and re-iterating thousands of AI experiments. On the flip side ML engineers are looking for distributed training, artifact management, and automated deployment for high performance

Read More
Book tickets View all speakers

We use cookies essential for this site to function well. Please click to help us improve its usefulness with additional cookies. Learn about our use of cookies in our Privacy Policy & Cookies Policy.

Show details