You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
I recently started a startup called Bespoke Labs. We are still in stealth and building our product! <br>
73
-
I was a Tech Lead in Google DeepMind and worked on LLM-based recommender models.
71
+
<h1class="mb-3">Recent Work</h1>
72
+
I started Bespoke Labs in 2024. We focus on RL environment curation and Data curation.<br>
73
+
<br>
74
+
Previously, I was a Tech Lead in Google DeepMind and worked on LLM-based recommender models. In 2021,
75
+
very few people were thinking about applying the bitter lesson to recommender systems.
76
+
I kickstarted a project to scale recommender models using LLMs, which resulted in
77
+
several research publications (and one best paper award) and many impactful launches across recommender systems such as YouTube.
74
78
<br>
75
-
76
-
I have been instrumental in scaling up our recommender models and delivered numerous launches that improve either the quality or the performance of recommender models. <br>
77
-
78
-
I work on all aspects of recommender models: infrastructure and productionization, research, and engagements. My work benefits surfaces such as YouTube.
<li>Etash Guha, Ryan Marten, Sedrick Keh, Negin Raoof, Georgios Smyrnis, .. <b>Maheswaran Sathiamoorthy</b>, Alex Dimakis, Ludwig Schmidt. "OpenThoughts: Data Recipes for Reasoning Models". <ahref="https://arxiv.org/abs/2506.04178">arXiv</a>. 2025</li>
94
+
<li>Yashar Deldjoo, Zhankui He, Julian McAuley, Anton Korikov, Scott Sanner, Arnau Ramisa, René Vidal, <b>Maheswaran Sathiamoorthy</b>, Atoosa Kasrizadeh, Silvia Milano, Francesco Ricci. "Recommendation with generative models". <ahref="https://arxiv.org/abs/2409.15173">arXiv</a>. 2024</li>
95
+
<li>Yashar Deldjoo, Zhankui He, Julian McAuley, Anton Korikov, Scott Sanner, Arnau Ramisa, René Vidal, <b>Maheswaran Sathiamoorthy</b>, Atoosa Kasirzadeh, Silvia Milano. "A review of modern recommender systems using generative models (gen-recsys)". Proceedings of the 30th ACM SIGKDD conference on Knowledge Discovery and Data Mining. 2024</li>
93
96
<li>Shashank Rajput, Nikhil Mehta, Anima Singh, Raghunandan H. Keshavan, Trung Vu, Lukasz Heldt, Lichan Hong, Yi Tay, Vinh Q. Tran, Jonah Samost, Maciej Kula, Ed H. Chi, <b>Maheswaran Sathiamoorthy</b>. "Recommender Systems with Generative Retrieval". <ahref="https://arxiv.org/abs/2305.05065">arxiv</a>. NeurIPS 2023</li>
94
97
<li>Jiaxi Tang, Yoel Drori, Daryl Chang, <b>Maheswaran Sathiamoorthy</b>, Justin Gilmer, Li Wei, Xinyang Yi, Lichan Hong, Ed H. Chi, "Improving Training Stability for Multitask Ranking Models in Recommender Systems". <ahref="https://arxiv.org/abs/2302.09178">arxiv</a>. KDD 2023, <b>Best Paper Award</b> in Data Science Track.</li>
95
98
<li>Wang-Cheng Kang, Jianmo Ni, Nikhil Mehta, <b>Maheswaran Sathiamoorthy</b>, Lichan Hong, Ed Chi, Derek Zhiyuan Cheng. "Do LLMs Understand User Preferences? Evaluating LLMs On User Rating Prediction". <ahref="https://arxiv.org/abs/2305.06474">arxiv</a>. Workshop on Personalized Generative AI, CIKM 2023.</li>
0 commit comments