Akshita Bhagia
Senior Research Engineer at AI2.
I am a Senior Research Engineer on the AllenNLP team, involved in R&D for natural language processing (NLP). Most recently, I have been working on the OLMo project, where I have contributed to
- pretraining dataset construction
- model training and inference
- evaluation tools and benchmark
I have also worked on open-source libraries such as allennlp, ai2-tango, etc.
I graduated with a Master’s degree in Computer Science from the University of Massachusetts Amherst in 2020, where I worked with Prof. Mohit Iyyer at the intersection of NLP and digital humanities.
Previously, I have worked at Cerebellum Capital (Summer 2019), and at InFoCusp (2015-2018), where I worked on building a data science platform.
news
| Dec 10, 2024 | Co-presented a tutorial on Data Preparation, Model Training, and Adaptation at Neurips 2024. |
|---|---|
| Aug 15, 2024 | Panelist on the Open Source AI panel at the AI/ALL Summit: Future of Equitable and Inclusive AI co-located with ACL 2024. |
| May 9, 2024 | Our work on the OLMo project won in the “Innovation of the Year” category at the Geekwire Awards! |
| Mar 26, 2024 | I presented our work on the OLMo project at a Databricks webinar. |
| Mar 4, 2024 | I was interviewed by Sam Charrington on the TWiML podcast about the OLMo project. |
| Feb 1, 2024 | We released the OLMo suite of models, along with the training code, the evaluation code, and the technical report. |
| Feb 1, 2024 | We released the DOLMa dataset and toolkit along with the technical report. |