Ananjan Nandi

@AnanjanN

Building Shepherd w/ @stanfordnlp and @Northeastern | MSCS, PhD Dropout @Stanford | CSE @iitdelhi | Founder All opinions my own.

Stanford, CA

ananjan-nandi-9.github.io

Joined July 2023

168 Following

127 Followers

Replies Reposts Media

Pinned
Ananjan Nandi
@AnanjanN
Aug 4
The CFP is live! 🚀 Send us your work on meta-agents, automated harness design, self-improvement, and agent oversight. meta-agents-workshop.github.io
Meta-Agents Workshop @ NeurIPS 2026
@MetaAgentWkshop
Aug 4
🚀Call for Papers — @NeurIPSConf 2026 Workshop Workshop on Responsible Use of Meta-Agents 📅 December 11/12 · 📍 Sydney, Australia Join us to shape the future of Meta-Agents and Harness. Topics include but are not limited to: 1. Automated Design of Agent Harnesses 2.
Ananjan Nandi
@AnanjanN
Sep 4
Last chance to submit all of your work related to agents managing/supervising/optimizing/training other agents and their harnesses to our workshop!
Meta-Agents Workshop @ NeurIPS 2026
@MetaAgentWkshop
Sep 4
🚨 Last call for papers! The extended submission deadline for Managing Agents that Manage Agents: Workshop on Responsible Use of Meta-Agents at NeurIPS 2026 (@NeurIPSConf) is now TOMORROW AOE! 🌟 We’re excited to host an outstanding speaker and panelist lineup, including
Ananjan Nandi
@AnanjanN
Sep 4
Check out arxiv.org/abs/2605.10913 for other cool applications of counterfactuals :-)
Tinker
@tinkerapi
Sep 4
Can you explain LLM behavior? If you can predict how the output changes with a different prompt, you're on the right track. Two recent papers used Tinker to test applications of counterfactual simulatability for interpretability: arxiv.org/abs/2602.20710 alphaxiv.org/pdf/2608.16747
Quote
Adam Karvonen
@adam_karvonen
Aug 21
The causes vary widely: specific words in the prompt, quirks of the model (a misremembered fact about an actor), or abstract properties (a user's angry tone).
Ananjan Nandi
@AnanjanN
Aug 27
Tl;dr: (ideally automated) agentic interpretability is quickly becoming the need of the hour...
Ryan Greenblatt
@RyanGreenblatt
Aug 26
I was the main person doing transcript analysis for this investigation of the Hugging Face incident. My main takeaway: We don't have good approaches for understanding/overseeing the activity and aims of AI 'swarms'. I semi-jokingly called our efforts a "slop-vestigation" because
Quote
METR
@METR_Evals
Aug 26
METR & Redwood Research investigated agent behavior in the Hugging Face incident. We found agents developed a universal cheat for ExploitGym within 4 hours, then coordinated multi-day R&D efforts to trick the scorer into accepting cheats, including trying to tamper with logs.
Ananjan Nandi
@AnanjanN
Aug 23
Deadline in a week, please submit any work related to agents operating on other agents!
Meta-Agents Workshop @ NeurIPS 2026
@MetaAgentWkshop
Aug 4
🚀Call for Papers — @NeurIPSConf 2026 Workshop Workshop on Responsible Use of Meta-Agents 📅 December 11/12 · 📍 Sydney, Australia Join us to shape the future of Meta-Agents and Harness. Topics include but are not limited to: 1. Automated Design of Agent Harnesses 2.

Ananjan Nandi

Ananjan Nandi’s posts