I am a Research Intern at the AI Institute, University of South Carolina, advised by Prof. Amit Sheth.
I'm an AI engineer and researcher working on making LLMs more reliable in practice. I study why models degrade during long-context generation, and I build lightweight diagnostics and retrain-free interventions that catch failures early. My work has been presented at ICML, AAAI, and NeurIPS. I'm currently doing an M.Sc. in Machine Learning at the University of Tübingen.
I am always open to discussing research or collaborations, feel free to reach out!
News
- Sept 2026 Paper accepted at NeurIPS 2026 Workshop LP4FM. See you in Sydney!
- Sept 2026 2 papers accepted at AAAI Fall Syposium 2026! Presenting in Arlington, VA.
- Apr 2026 Paper accepted at ICML 2026 — Cognitive Fatigue in Autoregressive Transformers: Formalization and Measurement.
- Jan 2026 Presented Chatsparent at AAAI 2026 in Singapore; served as a Student Scholar Volunteer.
- Oct 2025 Paper accepted at the AAAI 2026 Demonstration Track — Chatsparent: An Interactive System for Detecting and Mitigating Cognitive Fatigue in LLMs (28% acceptance rate).
Publications
-
1.
Cognitive Fatigue in Autoregressive Transformers: Formalization and Measurement ICML 2026Formalizes cognitive fatigue as a runtime state variable grounded in three token-level signals. Introduces the Fatigue Index with five explicit axioms; validated across nine models (1B–13B) on HotpotQA, TriviaQA, and SQuAD under long-context, positional, and precision stress conditions.
-
2.
Chatsparent: An Interactive System for Detecting and Mitigating Cognitive Fatigue in LLMs AAAI 2026 DemoAn interactive system that detects and mitigates cognitive fatigue in LLMs via token-level signals and retrain-free interventions, improving reliability and transparency during dialogue.
-
3.
Output Language Confusion under Multilingual Prompt Contamination NeurIPS 2026 Workshop LP4FMDeveloped MDI, a multilingual robustness evaluation protocol, and evaluated 5 LLMs across 40K trials, identifying script-switching and abstention as distinct failure modes and exposing systematic inflation in exact-match hallucination metrics.
-
4.
Context Matters: Evaluating LLM-Generated Knowledge Graph Schemas AAAI 2026 Fall SyposiumDesigned and evaluated LLM-generated KG schemas across four contextual regimes using intrinsic conformance metrics and extrinsic biomedical QA, achieving 78–80% edge conformance against a 79.9% expert-engineered baseline.
-
5.
Fatigue-Resistant Post-Training: Mitigating Cognitive Fatigue in Autoregressive Transformers via FI-Guided Optimization AAAI 2026 Fall SyposiumFatigue-Resistant Post-Training. Developed Fatigue Index-guided post-training methods for reducing Cognitive Fatigue (1,2) in autoregressive LLMs, using attention decay, entropy deviation, and representation drift across SFT, GRPO, and DAPO.
Experience
-
Research Intern AI Institute, University of South Carolina Apr 2025 – Present
- Advised by Prof. Amit Sheth (NCR Chair & Director, AIISC) and Vishal Pallagani.
- Designed a framework to detect and mitigate cognitive fatigue in LLMs using token-level signals and real-time interventions.
- First-authored publications at AAAI 2026 (Demo) and ICML 2026; researching long-context reliability, entropy collapse, and attention decay.
-
Research Collaborator University of Illinois Urbana-Champaign Nov 2025 – Mar 2026
- Studied politeness framing and reward leakage in LLMs across structured tasks and instruction-following settings.
- Performed mechanistic interpretability analysis including early-token probing and activation patching.
-
Generative AI Intern EY (Ernst & Young) Jan 2025 – Mar 2025
- Built a low-code platform for agentic AI workflows using modular DAGs, vector DBs, and LLM toolchains (OpenAI, FAISS).
- Implemented Celery–Redis task execution with production-grade scalability and semantic agent routing.
-
Software Developer Intern National Thermal Power Corporation Jul 2024 – Sep 2024
- Developed ASP.NET Core applications using MVC and Entity Framework for enterprise automation.
- Implemented secure authentication using JWT, Identity Framework, and Google reCAPTCHA.
Projects
↗
Research · LLM Interpretability
Cognitive Fatigue in LLMs
Formalizing cognitive fatigue as a measurable latent state in autoregressive transformers.
Fatigue Index, token-level signals, retrain-free mitigations.
↗
Industry · AI Engineering
Retrieval Drift Monitoring
RAG pipelines degrade silently with zero visibility. We build continuous live monitoring that
tests your pipeline against known answers, tracks chunk retrieval quality over time, detects performance drift, and alerts you instantly.
↗
Industry · Agentic AI
Intelligent Workflow Orchestration
A modular, DAG-based low-code platform for composing AI-augmented pipelines using LLM agents,
vector stores, and real-world data connectors. Built at EY.
↗
Community · Guide
Masters in Germany: A Complete Guide
A detailed guide for students navigating German MSc admissions, covering universities,
applications, funding, and life in Germany.
Honors & Service
- AAAI Student Volunteer Scholarship – USD 1,100 travel grant, AAAI 2026
- Student Scholar Volunteer – AAAI 2026, Singapore
- McKinsey Forward Program – selected for global initiative on problem-solving, business, and leadership
- Delegate – Harvard Project for Asian & International Relations (HPAIR), 2025
- Media Coverage – Times of India, Hindustan Times, Dwarka Parichay, Brainfeed
Education
-
M.Sc. Machine Learning IncomingUniversity of TübingenSep 2026 –
-
B.Tech. Computer Science EngineeringGuru Gobind Singh Indraprastha UniversityNov 2022 – Jun 2026