Aspiring Student-run Research Lab · Est. 2025

Asking the right questions about AI.

Viveka is an aspiring student-led research lab investigating the internal workings of large language models — mechanistic interpretability, reasoning, and the frontiers of AGI evaluation.

Viveka research team group photo
The Viveka Team 11 student researchers · 2025–2026
"Viveka" — Sanskrit for discernment, the ability to distinguish truth from illusion.
01 — About Us

A group built on curiosity and rigor.

Viveka is an aspiring student-run research lab focused on mechanistic interpretability and the evaluation of large language models. We are a group of undergraduate researchers who believe that understanding AI from the inside is one of the most important scientific challenges of our time.

We operate in year-long tenures, each with a focused research agenda, a dedicated team, and a commitment to impactful work. Our research is open, collaborative, and grounded in both theoretical depth and empirical rigor.

We welcome advisors, collaborators, and supporters who share our conviction.

01
Open research
All findings, code, and datasets are shared publicly.
02
Student-led, mentor-guided
Students drive the agenda. Advisors provide depth and direction. Project leads nurture the next set of undergraduate student research leads every tenure.
03
Mechanistic focus
We go beyond benchmarks — we want to know why models behave the way they do.
04
Impactful work
Our goal every tenure is to try and make our best contribution to the problem at hand.
02 — Tenures

Two focused chapters of research.

Tenure One
Viveka 1.0
2025 — 2026
Mechanistic Interpretability Hallucinations Two hop Factual Recall In-Context Learning Mealy markov models and transformers

Our inaugural tenure investigates the mechanistic underpinnings of LLM hallucinations, factual recall circuits, and the dynamics of in-context learning. Through probing, activation patching, causal tracing, training toy models in mathematical frameworks, we try to build rigorous explanations of model internals from the ground up.

Explore Viveka 1.0
Tenure Two
Viveka 2.0
2026 — 2027
ARC-AGI 2 Abstract Reasoning AGI Evaluation

Building on the interpretability foundation of V1.0, our second tenure turns toward frontier AGI evaluation. We interpret and advance performance on ARC-AGI 2, probing how (and whether) models generalise, reason abstractly, and compose solutions to genuinely novel problems.

Explore Viveka 2.0
03 — Advisors

Guided by those who've gone further.

NM
Dr. Nitish Mital
Research Mentor
Digital Catapult, London, UK
AC
Prof. Avishek Chatterjee
Research Mentor
Department of Electrical Engineering, IIT Madras, India
KM
Prof. Kaushik Mitra
Faculty Advisor, AI Club, CFI
Department of Electrical Engineering, IIT Madras, India
04 — Supporters

Made possible by generous support.

We are grateful for the compute, funding, and mentorship provided by our supporters. If you'd like to support Viveka, reach out.

ExceptionRaised!
Atomic Grant / GradCapital
Ankit Chaudhary