Modern IR Evaluation for RAG @ MILA
Presentation at MILA, Montreal, Quebec
CV
Here is my Academic CV in the PDF format (Last updated in March 2026):
Academic CV (Text Format)
Education
- Ph.D. in Computer Science, University of Waterloo, April 2026
- B.E. (Hons.) in Electronics & Instrumentation, BITS Pilani, 2018
Research Experience
- Fall 2024: Internship at Databricks
- Project: Building a code-based realistic RAG evaluation benchmark on niche topics.
- Supervisors: Omar Khattab, Sam Havens, Michael Carbin and Andrew Drozdov.
- Winter & Summer 2024 : Internship at Vectara
- Project: Building an effective multilingual RAG benchmark with heuristic & automatic features.
- Supervisors: Amin Ahmad and Suleman Kazi.
- Fall 2022 & Winter 2023: Internship at Google Research
- Project: Building an efficient multilingual embedding model using LLM-generated synthetic data.
- Supervisors: Daniel Cer and Jianmo Ni.
- Nov 2019 - Aug 2021: Research Assistant at UKP Lab, TU Darmstadt
- Project: Building a zero-shot heterogeneous benchmark and data-efficient sentence embedding models.
- Supervisors: Nils Reimers and Iryna Gurevych.
Work Experience
- Sep 2018 - Oct 2018: Data Scientist at KNOLSKAPE
- Project: Developed Krawler.ai, an deep text-based search engine built on Lucene.
- Manager: Chaithanya Yambari
Talks (List)
Modern IR Evaluation for RAG @ Hamel’s Eval Course
Presentation at Hamel's Eval Course, Virtual, Online
Beyond Models Rethinking Benchmarks, Data, and Evaluation for Retrieval-Augmented Generation. @ IISc Bangalore”
Presentation at Indian Institute of Science, Bangalore, India
RAG Benchmarks @ Weaviate Podcast
Podcast at Weaviate, Virtual, Online
Accelerating Multilingual RAG @ Micrsoft Research
Invited Talk at Microsoft Research, Bangalore, India
Heterogeneous IR Benchmarking @ IIIT Delhi
Invited Talk at IIIT-Delhi, Delhi, India
Heterogeneous IR Benchmarking @ IIT Delhi
Invited Talk at IIT-Delhi, Delhi, India
Advanced Information Retrieval @ Koç
Invited Talk at Koç University, Virtual (Online)
Heterogeneous Benchmarking @ Stanford
Invited Talk at Stanford University, California, USA
BEIR: An Open-Source Benchmark for Information Retrieval Systems @ OpenNLP
Invited Talk at OpenNLP Meetup (Deepset.ai), Virtual (Online)
Teaching
Service and leadership
- Shared-Task Lead Organizer on Retrieval Augmented Generation (RAG) Track at TREC 2024.
- Competition Lead Organizer on MIRACL at WSDM Cup 2023.
- Reviewer (*CL/NLP conferences): ACL Rolling Review: Oct-Nov (2021), Jan-Apr (2022)
- Reviewer (ML conferences): NeurIPS 2023.
- Reviewer (IR conferences): SIGIR 2023, ECIR 2024, NAACL 2024.