01 - About
Researcher, builder, occasional fund-raiser.
I'm an AI/ML engineer pursuing my M.Tech in Artificial Intelligence at IISc Bangalore, after a B.Sc. in Data Science & Applications at IIT Madras and a B.E. in Computer Science, completed in parallel. My thesis work investigates coverage-guaranteed conformal prediction for uncertainty inheritance in multi-agent LLM systems, and I build multimodal retrieval-augmented generation, machine unlearning, and applied GenAI systems - the kind that need to reason over messy, real-world data without falling over.
Outside the lab, I've run fundraising drives, led a 5,000-attendee fest's registration desk, and managed budgets for one of India's largest techno-cultural festivals. I like problems with a deadline and a leaderboard.
LocationBengaluru, India
FocusMulti-Agent LLMs · RAG · NLP
StatusM.Tech '27, IISc
TeachingTA, Linear Algebra, IISc
LanguagesPython · JavaScript
02 - Education
Academic profile
M.Tech, Artificial Intelligence
Indian Institute of Science
B.Sc., Data Science & Applications [🥇]
Indian Institute of Technology, Madras
B.E., Computer Science & Engineering
Easwari Engineering College (SRM)
03 - Skills
Tooling, by grade
AuCore AI / ML
Machine Learning
NLP
LLMs
RAG
Optimization
Multi-Agent Systems
AgFrameworks & Libraries
PyTorch
HuggingFace
Transformers
LangChain
Scikit-learn
XGBoost
SHAP
CLIP
YOLOv8
Pandas
NumPy
PtLanguages
Python
JavaScript
TiEngineering & Web
Flask
Vue.js
Celery
Redis
Streamlit
Git/GitHub
04 - Projects
Selected work
Uncertainty Quantification in Multi-Agent SystemsJul '26 – Present
M.Tech Thesis
- Investigating coverage-guaranteed conformal prediction for cross-agent uncertainty inheritance in dynamic DAG-structured multi-agent LLM systems, a gap unaddressed by existing frameworks (G-Designer, DySCo, UProp, DebUnc).
- Identified and formalized theoretical flaws in an existing e-process-based topology design framework, including an independence violation and a "groupthink" failure mode from correlated agent errors.
- Developing calibrated confidence scoring between connected agents and a method to compose scores across a multi-agent chain for a provable end-to-end accuracy guarantee.
Multimodal RAG PipelineAug '26 – Sep '26
Personal Project
- Implemented a BERT-style transformer text encoder from scratch in PyTorch, validated to cosine similarity 1.0 against a reference bge-base-en-v1.5 checkpoint, then fine-tuned via contrastive learning on 82K mined query–passage pairs, improving Recall@1 0.39→0.65.
- Architected a dual-path multimodal retriever over a 638-document PDF corpus, pairing the custom text encoder with frozen CLIP for images, recovering ~1,200 figures via a CLIP-based attachment gate at ~85% precision.
- Designed cross-encoder reranking for adaptive top-k selection, grounding a Qwen-VL backbone in retrieved text and figures to cut irrelevant context by ~30%.
Mixture of Experts for Recommendation SystemsSep '26
Course Project
- Built a self-attention Mixture-of-Experts recommender (MoSE), replacing LSTM experts with 4 parallel causal self-attention experts under a shared MMoE gate for joint next-item ranking and rating prediction on MovieLens.
- Achieved 0.282 Recall@10 / 0.163 NDCG@10 (~3× an LSTM-expert baseline) while cutting rating RMSE from 1.177 to 1.097.
- Diagnosed emergent expert specialization via gate-weight and attention analysis, consistent with MMoE routing dynamics reported in RecSys'19.
05 - Experience
Internships & teaching
Aug '26 – Present
Teaching Assistant, Linear Algebra (E0 298)
Indian Institute of Science, Bangalore
TA for E0 298 – Linear Algebra and Its Applications, current semester.
May '26 – Jul '26
Data Scientist Intern
IBM
Built an end-to-end AI pipeline over 600+ multimodal marketing assets to power a customer-behavior model; extracted 5,000+ topics via LLMs and clustered them into 60+ semantic personas; modeled engagement/revenue drivers with XGBoost and SHAP; shipped a conversational agent over a centralized content database for non-technical stakeholders.
May '25 – Jun '25
Software Engineer Intern
Simulation Software and Technology Pte Ltd
Built full-stack features across a JavaScript frontend and Python backend, designing and integrating REST APIs with end-to-end testing; trained and integrated a query-handling model into a production-facing interface.
Sep '24 – Nov '24
Research Intern
Vyoma Systems Pvt Ltd (IITM Research Park)
Automated a chip verification pipeline, integrating a large-scale instruction test-generation library with an LLM-based test-code generator in a coverage-driven feedback loop, eliminating manual test authoring.
Jan '23 – May '24
Teaching Assistant, Statistics
Indian Institute of Technology, Madras
Led weekly live sessions, developed graded activities and supplementary content, and evaluated student performance.
06 - Recognition
Publications & achievements
Published "Machine Unlearning in Recommendation Systems", IEEE ICTEST 2025 - indexed in IEEE Xplore.
IEEE
Top 100 Scholar, nationwide - Reliance Foundation Postgraduate Scholarship.
Wells Fargo Fellowship - awarded for academic merit.
Academic Excellence Award - B.Sc. Data Science program, IIT Madras.
Star of Rotaract - Rotary International, for outstanding social responsibility.
07 - Leadership
Positions of responsibility
Placement Representative
M.Tech AI, IISc · 2026–27
Point of contact between the M.Tech cohort and recruiters for the AI program's placement process.
Finance Head
Computer Society of India
Managed budgeting and financial planning for society events and initiatives.
Professional Service Chairwoman
Rotaract Club · 2023–24
Planned and led fundraising initiatives supporting prosthetic leg assistance.
Deputy Head, Finance & Operations
Paradox, IIT Madras · '23 & '24
Led budgeting, logistics, and vendor operations for IIT Madras' annual techno-cultural fest; previously Head of Registration (5,000+ attendees, 50+ volunteers) and QMS Coordinator.
Sports & Cultural Director
Rotaract Club · 2022–23
Organized events and raised ₹2L toward cleft-lip surgeries for children.