About

Hi, I am Prahaladh, a founding engineer at Circle-AI, where I am building AI agents for insurance. I am Engineering hire #2 for this company funded by Unusual Ventures, and I am primarily responsible for the backend and infra of our product.

I recently graduated from the School of Computer Science (SCS), CMU. I worked as a machine learning engineer at the Language Technologies Institute, CMU, advised by Prof. Chenyan Xiong and Prof. Mona Diab, where my primary role was to develop tools that aid multimodal RAG and deep research agents research.

I was also on the organizing committee of the NeurIPS competition: MMU-RAG: the Massive Multi-Modal User-Centric Retrieval-Augmented Generation Benchmark (Competition Website), where my primary role involved the system design and implementation of the live evaluation platform along with participant registration and leaderboard calculation mechanisms.

I was also a privacy engineer in the Software and Societal Systems department, CMU, where I worked under Professors Norman Sadeh, Lorrie Cranor and Hana Habib on developing comprehensive case studies for applying the UsersFirst framework. For my capstone project, I partnered with Meta's AI Risk and Governance team to investigate user privacy expectations and develop solutions for enhancing transparency in AI voice assistants.

Before joining CMU, I worked as a Software Engineer at Bank of America in their APAC Payments team and also worked as a Software Engineer intern at the middleware engineering division of RedHat.

Keywords: AI Agents, Multimodal RAG, Privacy Engineering, Federated Learning, Evaluation Platforms.

Want to chat? Send me an email at prahald92 at gmail dot com or a text on LinkedIn.

Timeline

Feb, 2026: RefusalBench Studio: An Interactive Platform for Generative Evaluation of LLM Selective Refusal (GitHub) (Demo)
Jan, 2026: Joined Circle-AI as Founding Forward Deployed Engineer.
Jan, 2026: Our Deep-Research-Comparator paper got accepted as demo paper in The Web Conference (WWW).
Dec, 2025: Graduated from CMU with a Master's in Privacy Engineering from the School of Computer Science, with an overall GPA of 3.92/4.33.
Sept, 2025: Started my Capstone Project with Meta on "Enhancing Transparency in AI Voice Assistants".
Aug, 2025: I am organizing a NeurIPS competition "MMU-RAG: the Massive Multi-Modal User-Centric Retrieval-Augmented Generation Benchmark". (Competition Website)
Jul, 2025: Published a Pre-Print "Deep Research Comparator: A Platform For Fine-grained Human Annotations of Deep Research Agents." (Arxiv)
Jan, 2025: Started as a Machine Learning Engineer building software tools that aid in Multimodal RAG and agentic evaluation as a collaboration between CMU LTI and Amazon AGI.
Jan, 2025: Started my independent study in Privacy threat modelling under Prof. Norman Sadeh, Lorrie Cranor and Hana Habib. (Framework Website)
Sept, 2024: Began my master's in Privacy Engineering from CMU. (LinkedIn Post)
Sept, 2023: Received the Arpit Jain Best Researcher Scholarship for our FL paper. (Website)
Aug, 2023: Filed my first patent through Bank Of America. (LinkedIn Post)
Sept, 2022: Joined Bank of America as Software Engineer in their APAC Payments team.
Jul, 2022: Published the Paper "Federated Learning for Colorectal Cancer Prediction." (IEEE Link)
Jan, 2022: Published the paper "Motion Pattern-based Crowd Scene Classification Using Histogram of Angular Deviations of Trajectories." (Journal Link)
Jan, 2022: Joined RedHat as Software Engineer intern in their middleware engineering division.
Oct, 2021: Joined a stealth startup as Federated Learning Engineer intern.
Mar, 2021: Joined Cloudanix (YC, 21) as a Software Engineer intern. (Cloudanix)
Nov, 2020: 2nd Place (National) in BRICS Future Skills Aerial Robotics competition, organised by WorldSkills Russia and the Moscow Regional Coordination Centre.
Jun, 2019 - Dec, 2020: Won various aeromodelling and Drone racing competitions as a part of AeroMIT.
Jun, 2019: Joined the Advanced Drone Research (now Autonomous Drone Research Subsystem) Subsystem in AeroMIT. (AeroMIT)

Research

Publications and research artifacts.

1

Deep Research Comparator: A Platform For Fine-grained Human Annotations of Deep Research Agents

Prahaladh Chandrahasan, Jiahe Jin, Zhihan Zhang, Tevin Wang, Andy Tang, Lucy Mo, Morteza Ziyadi, Leonardo F.R. Ribeiro, Zimeng Qiu, Markus Dreyer, Akari Asai, Chenyan Xiong

arXiv preprint arXiv:2507.05495, Jul. 2025.

2

Federated Learning for Colorectal Cancer Prediction

Yash Maurya, Prahaladh Chandrahasan, and Poornalatha G.

2022 IEEE 3rd Global Conference for Advancement in Technology (GCAT), pp. 1-5, Oct. 2022.

3

Motion Pattern-based Crowd Scene Classification Using Histogram of Angular Deviations of Trajectories

A. K. Pai, Prahaladh Chandrahasan, U. Raghavendra, and A. K. Karunakar.

The Visual Computer, vol. 39, no. 2, pp. 557-567, Jan. 2022.

Online Presence

Open Thoughts

2026.10.1 Some stuff about CMU Latest
Clicky