Abhimanyu Das
Google Principal Research Scientist leading structured-data models such as TimesFM; TimesFM paper first author
Abhimanyu Das is a Principal Research Scientist at Google, where Das leads efforts on predictive models for structured data, including the time-series and tabular foundation models TimesFM and TabFM. Das is first author of "A decoder-only foundation model for time-series forecasting", the paper behind TimesFM, and earlier worked as a researcher at Microsoft Research and Yahoo! Labs.12
Career
Das obtained a PhD in computer science from the University of Southern California under David Kempe, with research on algorithms for subset selection and sparse approximation. Before the PhD, Das spent several years in industry at Cortina Systems and Mahi Networks, working on software stacks for networking and embedded systems.1
Before Google, Das was a researcher at Microsoft Research and at Yahoo! Labs. At Google, Das is now a Principal Research Scientist heading work on predictive models for structured data.1
Research
Das's research spans machine learning, data mining and theoretical computer science. Early work with David Kempe connected submodularity and spectral analysis for subset selection (ICML 2011), and later papers covered display advertising, graph neural networks and bandits.1
Das's recent work centers on time-series forecasting: the TiDE encoder (2023), the decoder-only TimesFM foundation model (posted in 2023 and presented at ICML 2024), in-context fine-tuning for time-series foundation models (ICML 2025), and multimodal time-series forecasting (2026). Google Research's TimesFM-3 announcement names Das among the collaborators.1324
Work
TimesFMOpen weights
First author, with Weihao Kong, Rajat Sen and Yichen Zhou, of the 2023 paper introducing the decoder-only time-series foundation model behind TimesFM, which the TimesFM 3.0 model card cites.251
TimesFM 3.0Open weights
Named in the Google Research announcement of TimesFM-3 as one of the collaborators on the joint work behind the multivariate forecasting model.4
Long-term Forecasting with TiDE: Time-series Dense Encoder (external site: arxiv.org)Published research
First author of the 2023 TiDE paper, an encoder-decoder model for long-term time-series forecasting, published in TMLR.31
Career timeline
2011
Received a Distinguished Paper Award at ICML 2011 for work on subset selection with David Kempe.1
2014
Received a Best Paper Award at WSDM 2014 for work on hierarchical multitask learning in display advertising.1
2018
Received a Best Paper Honorable Mention at WWW 2018 for work on latency in ride and delivery services.1
Apr 2023
Posted the TiDE paper on long-term time-series forecasting as first author.3
Oct 2023
Posted the decoder-only time-series foundation model paper behind TimesFM as first author.2
Aug 2026
Named among the collaborators on TimesFM-3 in the Google Research announcement.4
Other documented roles
Roles the sources document but that are not confirmed as current.
- Researcher, Microsoft Research1
- Researcher, Yahoo! Labs1
Honors and recognition
Related people
More sourced profiles at the same organization, in the same field, or from the same school.
- Adam PaszkeAlso at Google
- Anima AnandkumarAI for science
- Ashish VaswaniAlso at Google Brain
- Alexei BaevskiAlso at Google
- Christopher BishopAlso at Microsoft Research
- Barret ZophAlso at Google Brain
- Amir HertzAlso at Google
- Daphne KollerAI for science
- Dani YogatamaAlso from USC
- Cordelia SchmidAlso at Google Research
- Graham TaylorAlso at Google
- Emanuel TodorovAlso from USC
Is this you?
Profiles cover public professional work only, and every line cites a source. If something is out of date, missing or wrong, send a correction with a link to an official page and we will update it. Suggest a correction
Add this badge to your website or GitHub profile README. It links back to this page. It is not an endorsement or a rating.
Markdown (for a README)
[](https://unitedstatesofamericasuperintelligence.com/people/abhimanyu-das/)
HTML
<a href="https://unitedstatesofamericasuperintelligence.com/people/abhimanyu-das/"><img src="https://unitedstatesofamericasuperintelligence.com/badges/profile.svg" alt="Abhimanyu Das — Profile on USASI" height="20"></a>
This profile covers public professional work only, from the sources below. It is not an endorsement or a ranking. To correct or remove it, use the correction form.