PeopleofSuperintelligence

Researcher

Alessandro Sordoni

Alessandro Sordoni is an AI researcher whose work spans information retrieval, dialogue generation, language models, and coding agents. At RALI, he pursued doctoral research on document and query representations and coauthored a Twitter-trained dialogue model in 2015. He coauthored the NewsQA machine-comprehension dataset in 2016 and a hierarchical model for dialogue generation in 2017. Ordered Neurons won ICLR's 2019 Best Paper Award, and he coauthored Deep Language Networks in 2023, debug-gym in 2025, and EvoLib in 2026. As of October 2026, he is an adjunct professor in Université de Montréal's Department of Computer Science and Operations Research.

Badges

Timeline

  1. 2015

    A Twitter dialogue model

    At RALI, he is a PhD student under Jian-Yun Nie in 2015. For his doctorate at Université de Montréal, he studies document and query representations for information retrieval. He coauthors a neural response-generation model trained end to end on Twitter conversations, with previous dialogue turns informing each response.

  2. 2016

    The NewsQA dataset

    In November 2016, he coauthors NewsQA, a machine-comprehension dataset with over 100,000 human-generated question-answer pairs drawn from more than 10,000 CNN articles.

  3. 2017

    A hierarchical dialogue model

    At Maluuba Inc., he coauthors A Hierarchical Latent Variable Encoder-Decoder Model for Generating Dialogues in 2017. Its stochastic variables span multiple time steps and help produce long, diverse responses while maintaining dialogue state.

  4. 2019

    Ordered Neurons wins ICLR

    With Yikang Shen, Shawn Tan, and Aaron Courville, he coauthors Ordered Neurons, a recurrent LSTM language model designed to model hierarchical tree structures. The paper wins ICLR's 2019 Best Paper Award.

  5. October 2023

    Deep Language Networks

    In October 2023, he coauthors Joint Prompt Optimization of Stacked LLMs using Variational Inference. Its Deep Language Network stacks two language models and learns a natural-language prompt for each layer.

  6. March 2025–April 2025

    Debug-gym for coding agents

    In March 2025, he coauthors debug-gym, a text-based environment where LLM agents interactively explore codebases while debugging. In April 2025, he is a Senior Principal Research Manager at Microsoft Research.

  7. May 2026

    EvoLib for test-time learning

    In May 2026, he coauthors Test-Time Learning with an Evolving Library. Its EvoLib framework lets language models accumulate and reuse knowledge across problem instances without parameter updates or external supervision.

  8. As of October 2026

    Adjunct professor at UdeM

    As of October 2026, he is an adjunct professor in Université de Montréal's Department of Computer Science and Operations Research.

Career

  1. NowUniversité de MontréalProfessor
  2. From 2025Microsoft ResearchSenior Principal Research Manager
  3. From 2017Maluuba Inc.

Education

  1. From 2015RALIPhD

From the biography.

  1. The median h-index is 48 at universities and 14 at frontier AI labsIncludes Université de Montréal. The median h-index is 48 at universities and 14 at frontier AI labs. We compare publishing careers, recent citations, dual appointments and job moves across 2,420 university researchers and 308 frontier-lab researchers.
  2. 20% of people at companies who shape AI also hold a university role, led by Google and Stanford UniversityIncludes Université de Montréal. We estimate that 539 of the 6,766 people with a known current organization (8%) hold both a university role and a company role, including 20% of people at companies and 13% of people at universities.
  3. Of 943 people shaping AI who worked at Google, 446 work there nowIncludes Université de Montréal. Our estimates of where the 943 people shaping AI who worked at Google and Google DeepMind work now: 446 at Google and 452 across 278 other organizations.
  4. The Massachusetts Institute of Technology trained or employed 44 of the 187 AI major-prize winners, more than any other universityIncludes Université de Montréal. The Massachusetts Institute of Technology trained or employed 44 of the 187 AI major-prize winners, more than any other university. We compare doctorate and 5+ year employment links, career timing and current workplaces for 187 major-prize and 1,111 best-paper winners.

Colleagues

People with dated positions at the same organization in the same years.

Sources and credits

Photo: We have no published photo of Alessandro Sordoni yet. Send us a free-licensed photo.

  1. A Neural Network Approach to Context-Sensitive Generation of Conversational Responses - Microsoft Research microsoft.com. Checked 21 October 2026.
  2. Lisa_teatalk - mailman.iro.umontreal.ca mailman.iro.umontreal.ca. Checked 21 October 2026.
  3. Alessandro Sordoni | Mila mila.quebec. Checked 21 October 2026.
  4. NewsQA: A Machine Comprehension Dataset - Microsoft Research microsoft.com. Checked 21 October 2026.
  5. A Hierarchical Latent Variable Encoder-Decoder Model for Generating Dialogues | Proceedings of the AAAI Conference on Artificial Intelligence ojs.aaai.org. Checked 21 October 2026.
  6. Best Paper Award iclr.cc. Checked 21 October 2026.
  7. Brief Report: Ordered Neurons: Integrating Tree Structures into Recurrent Neural Networks - Microsoft Research microsoft.com. Checked 21 October 2026.
  8. Ordered Neurons: Integrating Tree Structures into Recurrent Neural Networks arxiv.org. Checked 21 October 2026.
  9. Joint Prompt Optimization of Stacked LLMs using Variational Inference - Microsoft Research microsoft.com. Checked 21 October 2026.
  10. debug-gym: A Text-Based Environment for Interactive Debugging arxiv.org. Checked 21 October 2026.
  11. Debug-gym: an environment for AI coding tools to learn how to debug code like programmers - Microsoft Research microsoft.com. Checked 21 October 2026.
  12. Test-Time Learning with an Evolving Library arxiv.org. Checked 21 October 2026.
  13. EvoLib: Turning experience into evolving knowledge - Microsoft Research microsoft.com. Checked 21 October 2026.
  14. Alessandro SORDONI - Département d'informatique et de recherche opérationnelle - Université de Montréal diro.umontreal.ca. Checked 21 October 2026.