Sagar Joglekar

Sagar Joglekar, Ph.D.

Machine Learning Scientist

Intercom

I am a Staff Machine Learning Scientist with over 10 years of experience bridging cutting-edge academic research and high-impact industrial applications. My expertise spans deep learning, computer vision, natural language processing, and complex networks, with a focus on developing novel methods that quantify human behavior from multimodal data.

Currently, I work as a Staff Machine Learning Scientist at Intercom's A.I. Group, where I lead initiatives that bring cutting-edge academic research into production systems. My work focuses on advancing customer support solutions through rigorous A/B testing and the development of scalable ML architectures.

I completed my Ph.D. in Computer Science at King's College London as a King's India Scholar, under the supervision of Dr. Nishanth Sastry. My doctoral research focused on quantifying signals from human behaviors driven by subjective perceptions, using data and AI to improve the world around us.

Throughout my career at organizations including Nokia Bell Labs, Expedia Group, and Citrix Systems, I have authored over 40 scientific publications in top-tier venues (Nature, WWW, ICWSM, CSCW), hold multiple patents, and have built both R&D solutions and production systems that solve real-world problems at scale.

About

Education

  • Ph.D. in Computer Science
    King's College, London, U.K.
    2015 – 2019
  • M.S. in Signals Processing and Networks
    University of California, Santa Barbara, U.S.A.
    2010 – 2012
  • B.Eng in Electronics Engineering
    University of Pune, India
    2004 – 2008

Research Interests

  • Complex Networks
  • Computational Social Dynamics
  • Representation Learning
  • Deep Learning
  • Natural Language Processing
  • Urban Informatics
  • AI Agents

Experience

Staff Machine Learning Scientist
Intercom
London, UK · Oct 2023 – Present

Help research and build the next generation customer engagement software powered by AI.

Machine Learning Scientist
Expedia Group
London, UK · Jul 2021 – Sep 2023

Developed machine learning algorithms and services for curation, ranking, and retrieval of content across Expedia Group products.

Research Scientist
Nokia Bell Labs
Cambridge, UK · Jun 2019 – Jul 2021

Worked on machine learning and statistical inference with a focus on healthcare and urbanism, mathematically modelling offline phenomena through online data.

Head of Research
Firedrop.ai
London, UK · Apr 2016 – Jun 2017

Research and Development. Advisor for all things ML.

Senior Software Engineer
Citrix Systems
Santa Barbara, CA, USA · Feb 2012 – Sep 2015

Development of distributed media encoder/decoder protocols for client/server side libraries in Citrix conferencing products (GoToMeeting, GoToAssist, GoToWebinar).

Research Engineer
Infosys Research Labs
Bangalore/Pune, India · Jul 2008 – Jul 2010

Research, prototyping, and development in state-of-the-art DRM, multimedia security and delivery systems.

Publications

Vitamin N: Benefits of Different Forms of Public Greenery for Urban Health Preprint
Sanja Scepanovic, Sagar Joglekar, Stephen Law, Daniele Quercia, Ke Zhou, Alice Battiston et al.
arXiv, 2025
A Graph-Based Context-Aware Model to Understand Online Conversations Journal
Vibhor Agarwal, Anthony P. Young, Sagar Joglekar, Nishanth Sastry
ACM Transactions on the Web, 2023
Our Nudges, Our Selves: Tailoring Mobile User Engagement Using Personality Conference
Nima Jamalian, Marios Constantinides, Sagar Joglekar, Xueni Pan, Daniele Quercia
LNCS, 2023
MedSat: A Public Health Dataset for England Featuring Medical Prescriptions and Satellite Imagery Conference
Laura Giustarini, Sagar Joglekar, Cristiano Nattero, Ivica Obadic, Daniele Quercia, Sanja Scepanovic et al.
NeurIPS Datasets Track, 2023
Multidimensional tie strength and economic development Selected Journal
Luca Maria Aiello, Sagar Joglekar, Daniele Quercia
Scientific Reports, 2022
GraphNLI: A Graph-based Natural Language Inference Model for Polarity Prediction in Online Debates Conference
Vibhor Agarwal, Sagar Joglekar, Anthony P Young, Nishanth Sastry
WWW 2022
Predicting Meeting Success With Nuanced Emotions Selected Journal
Ke Zhou, Marios Constantinides, Sagar Joglekar, Daniele Quercia
IEEE Pervasive Computing, 2022
On the adoption of e-moped sharing systems Journal
Stefano Fiorini, Michele Ciavotta, Sagar Joglekar, Sanja Scepanovic, Daniele Quercia
EPJ Data Science, 2022
Modelling online debates with argumentation theory Journal
Anthony P. Young, Sagar Joglekar, Vibhor Agarwal, Nishanth Sastry
ACM SIGWEB Newsletter, 2022
The Role of Different Types of Conversations for Meeting Success Selected Journal
Ke Zhou, Marios Constantinides, Luca Maria Aiello, Sagar Joglekar, Daniele Quercia
IEEE Pervasive Computing, 2021
Imagine a Walkable City: Physical activity and urban imageability across 19 major cities Conference
Marios Constantinides, Sagar Joglekar, Sanja Scepanovic, Daniele Quercia
EPJ Data Science, 2021
Retrofitting Meetings for Psychological Safety Conference
Marios Constantinides, Sagar Joglekar, Daniele Quercia
CSCW 2021
How epidemic psychology works on Twitter: evolution of responses to the COVID-19 pandemic in the US Journal
Luca Maria Aiello, Daniele Quercia, Ke Zhou, Marios Constantinides, Sanja Scepanovic, Sagar Joglekar
Humanities & Social Sciences Communications (Nature), 2021
KAIROS: Talking Heads and Moving Bodies for Successful Meetings Conference
Jun-Ho Choi, Marios Constantinides, Sagar Joglekar, Daniele Quercia
HotMobile 2021
Jane Jacobs in the Sky: Predicting Urban Vitality with Open Satellite Data Conference
Sanja Scepanovic, Sagar Joglekar, Stephen Law, Daniele Quercia
CSCW 2021
The Healthy States of America: Creating a Health Taxonomy with Social Media Conference
Sanja Scepanovic, Luca Maria Aiello, Ke Zhou, Sagar Joglekar, Daniele Quercia
ICWSM 2021
Characterising User Content on a Multi-lingual Social Network Selected Conference
Pushkal Agarwal, Kiran Garimella, Sagar Joglekar, Nishanth Sastry, Gareth Tyson
ICWSM 2020
Analysing Meso and Macro conversation structures in an online suicide support forum Selected Journal
Sagar Joglekar, Sumithra Velupillai, Rina Dutta, Nishanth Sastry
arXiv preprint, 2020
Who Has the Last Word? Understanding How to Sample Online Discussions Journal
Boschi G., Young A.P., Sagar Joglekar, Chiara Cammarota, Sastry N.
ACM TWEB, 2020
Ranking comment sorting policies in online debates Conference
Young A.P., Sagar Joglekar, Boschi G., Sastry N.
Argument & Computation, 2020
Stop Tracking Me Bro! Differential Tracking Of User Demographics On Hyper-partisan Websites Conference
Pushkal Agarwal, Sagar Joglekar, Panagiotis Papadopoulos, Nishanth Sastry, Nicolas Kourtellis
WWW 2020
FaceLift: a transparent deep learning framework to beautify urban scenes Selected Journal
Sagar Joglekar, Daniele Quercia, Miriam Redi, Luca Aiello, Tobias Kauer, Nishanth Sastry
Royal Society Open Science, 2020
Humane Visual AI: Telling the Stories Behind a Medical Condition Conference
Wonyoung So, Edyta P. Bogucka, Sanja Scepanovic, Sagar Joglekar, Ke Zhou, Daniele Quercia
IEEE VIS 2020
MeetCues: Supporting Online Meetings Experience Conference
Aseniero (Bon Adriel), Marios Constantinides, Sagar Joglekar, Ke Zhou, Daniele Quercia
IEEE VIS 2020
Challenges in the Decentralised Web: The Mastodon Case Selected Conference
Aravindh Raman, Sagar Joglekar, Emiliano De Cristofaro, Nishanth Sastry, Gareth Tyson
IMC 2019
Mapping and Visualizing Deep-Learning Urban Beautification Selected Journal
Tobias Kauer, Sagar Joglekar, Daniele Quercia, Miriam Redi, Luca Aiello
IEEE Computer Graphics & Applications, 2018
How do online communities of people with long-term conditions function and evolve? Selected Journal
Sagar Joglekar, Nishanth Sastry, Anna Di Simoni, Pietro Panzarasa et al.
JMIR, 2018
Illuminating an ecosystem of partisan websites Conference
Shweta Bhat, Sagar Joglekar, Shehar Bano, Nishanth Sastry
WWW 2018
Like at First Sight: Understanding User Engagement with the World of Microvideos Selected Conference
Sagar Joglekar, Miriam Redi, Nishanth Sastry
SocInfo 2017
Structure and dynamics of online patients' communities: the case of Asthma UK and BLF online fora Conference
Anna de Simoni, Sagar Joglekar, Nishanth Sastry, Pietro Panzarasa et al.
ERS 2017
Robust transcoding resistant watermarking for H.264 standard Journal
Rohit Nair, Vijayraghavan V, Sagar Joglekar et al.
Multimedia Tools & Applications, 2014
An Innovative System for Remote and Automated Testing of Mobile Phone Applications Conference
Karthikeyan Dhanapal, Sagar Joglekar, Arun Somasundara et al.
SRII 2012

Posts

Work update: Applied research at Intercom
On the boundaries of current AI technology — David vs Goliath in LLMs, Agency-Control-Reliability tradeoffs, and model alignment.
Actions and their (lack of) consequences
On the role of outrage in tribal social balance — and what it means for the metaverse.
Sensing the real world through NHS prescriptions data
Using NHS prescriptions data to create proxies for population health across geographic units.
The cost of being on the fence
Reflections on transitioning from blue-sky research at Bell Labs to product-oriented data science at Expedia Group.
Press cover for FaceLift
FaceLift published in Royal Society Open Science — a deep learning framework to beautify urban scenes. Covered by Fast Company and Cosmos Magazine.

Projects

Talks

CTR Workshop 2018
Center for Telecommunications Research Workshop
May 2018 · King's College London
SocInfo 2017
Sep 2017 · Oxford, UK