My research investigates machine learning ecosystems where an ML model (such as a language model) interacts with humans, companies, and other ML models. My goal is to steer how multi-agent interactions shape performance trends and societal outcomes. To tackle this, my work traces ecosystem-level outcomes back to the incentives of interacting agents and to the ML pipeline for training models. For example, some of my work studies scaling laws under competition, feedback loops with LLM agents, and the market of companies training LLMs.
I recently received a PhD in Computer Science from UC Berkeley, where I was advised by Michael I. Jordan and Jacob Steinhardt. My PhD research was supported by an Open Philanthropy AI Fellowship and a PD Soros Fellowship. During my PhD, I interned at Microsoft Research in the EconCS Group and in the ML & Optimization Group. I went to Harvard for undergrad, where I was advised by Jelani Nelson.
Here's my CV and my contact info.
Group
I'll be recruiting PhD students in the December 2025 application cycle. I'm excited to work with students who have a strong technical skillset (either theoretical, empirical, or a mixture of the two) and are interested in tackling conceptual questions that emerge in machine learning ecosystems. If you are interested in joining my group, please apply to the CIS PhD program at Penn and list my name on your application.I'll also be recruiting postdocs; if you are interested in working with me, please email me with a 2-3 sentence description of your research interests and two of your representative papers.
If you're currently a PhD student at Penn CIS and are interested in my research, feel free to email me!
Papers
(* denotes equal contribution, α-β denotes alphabetical ordering)Preprints
- Safety vs. Performance: How Multi-Objective Learning Reduces Barriers to Market Entry
Meena Jagadeesan, Michael I. Jordan, and Jacob Steinhardt.
[arXiv] [slides] - Flattening Supply Chains: When do Technology Improvements lead to Disintermediation?
(α-β) S. Nageeb Ali, Nicole Immorlica, Meena Jagadeesan, and Brendan Lucier.
[arXiv] - Incentivizing High-Quality Content in Online Recommender Systems
Xinyan Hu*, Meena Jagadeesan*, Michael I. Jordan, and Jacob Steinhardt.
[arXiv]
- Accounting for AI and Users Shaping One Another: The Role of Mathematical Models
(α-β) Sarah Dean, Evan Dong, Meena Jagadeesan, and Liu Leqi.
TMLR, 2025. Position Paper. (Presented under a different title at AAAI 2024 Workshop on Recommendation Ecosystems.)
[arXiv] - Impact of Decentralized Learning on Player Utilities in Stackelberg Games
(α-β) Kate Donahue, Nicole Immorlica, Meena Jagadeesan, Brendan Lucier, and Aleksandrs Slivkins.
ICML 2024. (Presented at ESIF-AIML 2024 and EC 2024 Workshop on Foundation Models and Game Theory.)
[paper] [slides] [poster] - Feedback Loops With Language Models Drive In-Context Reward Hacking
Alexander Pan, Erik Jones, Meena Jagadeesan, and Jacob Steinhardt.
ICML 2024.
[paper] [poster] - Clickbait vs. Quality: How Engagement-Based Optimization Shapes the Content Landscape in Online Platforms
(α-β) Nicole Immorlica, Meena Jagadeesan, and Brendan Lucier.
The Web Conference (WWW) 2024.
[paper] [talk] [slides] [poster] - Can Probabilistic Feedback Drive User Impacts in Online Platforms?
(α-β) Jessica Dai, Bailey Flanigan, Nika Haghtalab, Meena Jagadeesan, and Chara Podimata
AISTATS 2024.
[paper] [poster]
- Improved Bayes Risk Can Yield Reduced Social Welfare Under Competition
Meena Jagadeesan, Michael I. Jordan, Jacob Steinhardt, and Nika Haghtalab.
NeurIPS 2023. (Presented at ESIF-AIML 2024.)
[paper] [talk] [slides] [poster] - Supply-Side Equilibria in Recommender Systems
Meena Jagadeesan, Nikhil Garg, and Jacob Steinhardt.
NeurIPS 2023. (Presented at FORC 2024.)
[paper] [talk] [slides] [poster] - Competition, Alignment, and Equilibria in Digital Marketplaces
Meena Jagadeesan, Michael I. Jordan, and Nika Haghtalab.
AAAI 2023.
[paper] [slides] [poster] - Performative Power
(α-β) Moritz Hardt, Meena Jagadeesan, and Celestine Mendler-Dünner.
NeurIPS 2022.
[paper] [talk] [slides] [poster] [MAIEI blog] - Regret Minimization with Performative Feedback
Meena Jagadeesan, Tijana Zrnic, and Celestine Mendler-Dünner.
ICML 2022. (Presented at FORC 2022.)
[paper] [slides] [talk] - Inductive Bias of Multi-Channel Linear Convolutional Networks with Bounded Weight Norm
Meena Jagadeesan, Ilya Razenshteyn, and Suriya Gunasekar.
COLT 2022. (Presented at ICML 2021 Workshop on Overparameterization.)
[paper] [slides] [talk at MIT] - Individual Fairness in Advertising Auctions through Inverse Proportionality
(α-β) Shuchi Chawla and Meena Jagadeesan.
ITCS 2022. (Presented at FORC 2021.)
[paper] [talk] [slides] - Learning Equilibria in Matching Markets from Bandit Feedback
Meena Jagadeesan*, Alexander Wei*, Yixin Wang, Michael I. Jordan, and Jacob Steinhardt.
NeurIPS 2021. Spotlight Presentation.
Full version in Journal of the ACM.
[paper] [poster] [slides] - Alternative Microfoundations for Strategic Classification
Meena Jagadeesan, Celestine Mendler-Dünner, and Moritz Hardt.
ICML 2021. (Presented at NeurIPS 2021 Workshop on Learning in Presence of Strategic Behavior.)
[paper] [talk] [slides] [poster] - Cosine: A Cloud-Cost Optimized Self-Designing Key-Value Storage Engine
Subarna Chatterjee, Meena Jagadeesan, Wilson Qin, and Stratos Idreos.
VLDB 2021.
[paper]
- Multi-Category Fairness in Sponsored Search Auctions
Christina Ilvento*, Meena Jagadeesan*, and Shuchi Chawla.
ACM FAT* 2020. (Presented at EC 2019 Workshop on Mechanism Design for Social Good.)
[paper] [slides] [talk] - Individual Fairness in Pipelines
(α-β) Cynthia Dwork, Christina Ilvento, and Meena Jagadeesan.
FORC 2020.
[paper] [full version] [slides from reading group] - Understanding Sparse JL for Feature Hashing
Meena Jagadeesan.
NeurIPS 2019. Oral Presentation.
[paper] [full version] [talk] [slides] [coverage] - Simple Analysis of Sparse, Sign-Consistent JL
Meena Jagadeesan.
RANDOM 2019.
[paper] [full version] [slides] - Varying the Number of Signals in Matching Markets
Meena Jagadeesan* and Alexander Wei*.
WINE 2018. (Presented at EC 2018 Workshop on Frontiers of Market Design.)
[paper] [full version] - Dyson's partition ranks and their multiplicative extensions
(α-β) Elaine Hou and Meena Jagadeesan.
The Ramanujan Journal, 2018.
[paper] - Mobius Polynomials of Face Posets of Convex Polytopes
Meena Jagadeesan and Susan Durst.
Communications in Algebra, 2016.
[paper]
- From Worst-Case to Average-Case Analysis: Accurate Latency Predictions for Key-Value Storage Engines
Meena Jagadeesan* and Garrett Tanzer*.
SIGMOD 2020 (2-Page Extended Abstract). Winner of SIGMOD SRC.
[short paper]
Theses
- Steering Machine Learning Ecosystems of Interacting Agents
Meena Jagadeesan.
PhD thesis from the CS Department at UC Berkeley, 2025.
[thesis] - The Performance of Johnson-Lindenstrauss Transforms: Beyond the Classical Framework
Meena Jagadeesan.
Undergraduate thesis from Harvard University, 2020. Awarded Hoopes Prize.
[thesis]
Selected Awards
- Open Philanthropy AI Fellowship (2021-2025)
- PD Soros Fellowship for New Americans (2020-2022)
- Berkeley Fellowship (2020-2023)
- CRA Outstanding Undergraduate Researcher Award (2020)
- Siebel Scholar (2019-2020)
- Barry Goldwater Scholar (2018-2020)
- Intel STS 2nd Place Prize in Basic Research (2016)
Contact Info
Email: mjagadeesan [at] berkeley [dot] edu[Google Scholar] [LinkedIn] [Twitter]