Skip to content
View ajithksenthil's full-sized avatar

Block or report ajithksenthil

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Maximum 250 characters. Please don't include any personal information such as legal names or email addresses. Markdown supported. This note will be visible to only you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
ajithksenthil/README.md

Ajith Senthil

PhD Student @ USC SAND Lab | Founder @ FairQuanta | Computational Models of Mind & Behavior


About Me

I'm a PhD student at the University of Southern California's SAND Lab, where I study behavior and computational models of personality using deep learning. My work sits at the intersection of AI, cognitive science, and human-centered design—driven by a core question: how can we build systems that truly understand human behavior?

This question has guided my journey from studying linguistics and brain science to founding FairQuanta, a research lab and human-centered design startup focused on context and human factors engineering.

Our Story

My research trajectory has followed a consistent thread: understanding the computational principles underlying human cognition and personality.

It began with dual degrees in Computer Science + Linguistics and Brain & Cognitive Science at UIUC, where I first encountered the power of combining computational methods with cognitive theory. A Master's in Psychological Science (with a Statistics minor) deepened my quantitative toolkit. Now, as a PhD student at USC, I'm developing deep learning approaches to model personality and behavior at scale.

Along the way, I co-founded the ByCog Research Group to explore Computational Psychodynamics—modeling personality through active inference. This work, combined with my contributions to the Active Inference Institute and Learning Language Lab, laid the foundation for FairQuanta: a venture dedicated to applying these insights to real-world human-centered systems.

Current Work

FairQuantaFounder & Research Lead Human-centered design startup and research lab focused on context and human factors engineering. We bridge computational models of cognition with practical system design.

USC SAND LabPhD Researcher Developing deep learning models of personality and behavior under the supervision of leading researchers in computational social science.

ByCog Research GroupCo-Founder Pioneering Computational Psychodynamics: process-based approaches to modeling cognition and personality using active inference.

Active Inference InstituteResearcher Building frameworks for interpreting human behavior and cognition through the lens of active inference.

Additional Roles

  • OpenPolitica — CEO & AI/ML Team Lead | Fostering democratic participation through AI-driven civic tools
  • UniversumX — Project Manager & ML Developer | Multi-university initiative (UIUC, UCLA, USC) developing deep learning models of perception with BCIs for AI alignment
  • Learning Language Lab — Researcher | Exploring linguistic and visual cognition

Technical Expertise

Research Methods: Bayesian Statistics, Statistical Learning, Mathematical Statistics, Regression Analysis, Deep Learning, Active Inference

Languages: Python, C++, Java, C#, R, C, PHP, SQL, JavaScript

ML/AI Stack: PyTorch, scikit-learn, HuggingFace Transformers, OpenAI API, LangChain, ChromaDB, NLTK, SpaCy

Tools: Git, Docker, Jupyter, Mathematica, Unity3D

Publications & Presentations

  • Senthil, A. (In preparation) Computational Psychodynamics: Process-Based Approach to Modeling Cognition and Personality Using Active Inference
  • Senthil, A. (2023) Computational Psychodynamics: Process-based approach to modeling personality [Poster presentation]

Connect

LinkedIn · Portfolio


Outside of research, I enjoy reading AI papers, watching TV shows, and mentoring future AI enthusiasts.

Pinned Loading

  1. MultiAgentRoadHazardAnomalyDetection MultiAgentRoadHazardAnomalyDetection Public

    Multi-Agent Robot Learning algorithm using Deep Active Inference (DAI) for road hazard anomaly detection and Soft Actor Critic decomposed for multi-agent settings (mSAC)

    Python 1

  2. UniversumX/Universum UniversumX/Universum Public

    Modeling Perception using Curriculum Learning, Transfer Learning to make incremental steps towards a generalizable model of perception and its AI applications

    Python 2 5

  3. PolicyWeb PolicyWeb Public

    AI Powered Tool for Democracy and Policy Making

    Python 1

  4. AttachmentBot AttachmentBot Public

    Chatbot driven conversations and predicting attachment scores from chat transcripts using embedding analysis + additional feature extraction and regression

    Python