Building AI agents that are trustworthy by design and aligned with the people they serve.
Hi, I'm Pranav (PRAH-nuv). I work on agentic AI: deep research and autoresearch, agent pipelines, agent-to-agent communication, and the simulated users we test agents on, so they earn human trust before they reach real people.
Taught ~200 PhD students at the Machine Learning Summer School how to rigorously measure, benchmark, and stress-test agentic systems as they take on complex, real-world tasks.
June 23, 2026 · Program ↗ Tutorial · KDD 2026 · Jeju, KoreaA tutorial on what changes when agents leave the lab, co-presented with G.H. Yang, H. Sedghamiz, E. Santus, V. Dibia, and I. Baldini.
August 2026 · Tutorial site ↗Every star is a paper, placed between the domains it bridges: deep research, agent pipelines, agent-to-agent communication, AI simulation, and human-centered AI, with Trust & Safety at the core because it runs through all of them. Drag to spin it, hover or tap a star, or pick a domain.
I study what AI agents do once they leave the lab (what they cite, how they talk to each other, who they simulate) and build the evaluations that tell us when to trust them.
I'm a Research Scientist at Salesforce AI Research on the Interactive AI team. Most of my work sits where agentic systems meet trust and safety. In deep research and autoresearch, I audit answer engines and deep research agents for unsupported claims and one-sided sourcing (DeepTRACE, ICLR 2026; Search Engines in the AI Era, FAccT 2025). I also help build enterprise deep research systems that know when they have enough evidence to stop (ACL 2026) and write reports that stay faithful to their sources, and I train reasoning models to propose diverse, grounded scientific ideas (Diverse Scientific Ideas, 2026).
A second thread is AI simulation and persona alignment. LLM personas increasingly stand in for real users, so I ask who those personas actually represent (A Tale of Two Identities, AAAI 2026), how to ground them in social science (InterviewSim, COLM 2026), and how AI companions drift out of character over long conversations. I also study agent-to-agent communication: how persuasion and beliefs cascade through LLM agent networks (Belief Cascades) and how images sway vision-language agents (MMPersuade and Seeing is Believing, EMNLP 2026).
All of it is human-centered AI at heart, and it grows out of my Ph.D. in Informatics at Penn State with Dr. Shomir Wilson, where I studied sociodemographic bias in NLP across disability, nationality, race, and culture. That work earned an Outstanding Paper Award at EMNLP 2023, a Best Short Paper Award at TrustNLP, a DEI Recognition Award at CSCW 2024, and an Honorable Mention for Penn State's Best AI Dissertation. I serve as an Area Chair for ACL and EMNLP and on the program committees of FAccT and AIES.
Thanks for visiting! Here are a few good reasons to get in touch. The list is probably incomplete.
You're a researcher
Questions about a paper, an idea you'd like to brainstorm or debug, or a possible collaboration on agent evaluation, deep research, personas, or responsible NLP.
You're a student
Thinking about a Ph.D. in NLP or HCI, or about doing research in industry? I'm glad to share what I've learned, especially if you're making the jump from engineering like I did.
You're a journalist
Looking for a technical perspective on AI search and deep research, LLM personas, or bias in language technology. My work has appeared in New Scientist, Fast Company, and The Hill.
You're building agents
Shipping an agent or LLM feature and want to think through evaluation, safety, or user simulation before it meets real users.
You're organizing something
A talk, panel, tutorial, or guest lecture on trustworthy agents or sociotechnical NLP.
You're in the Bay Area
Up for coffee and a conversation about NLP, HCI, or responsible AI with like-minded folks? Let's meet.
When in doubt, reach out!