Arnav Dhariya
I'm an undergraduate researcher at the University of California, Irvine, graduating June 2027 with a B.S. in Computer Science. My work spans data privacy and deletion, distributed systems for vector search and ML workloads, and agentic AI infrastructure. I work across research groups advised by Prof. Sharad Mehrotra, Prof. Sangeetha Abdu Jyothi, Prof. G.P. Li, Prof. Wayne Hayes, and Prof. Glenn Healy.
I'm an undergraduate at the University of California, Irvine, graduating in June 2027 with a B.S. in Computer Science. Dean's Honor List, 6 quarters.
My research sits at the intersection of database systems, distributed infrastructure, and agentic AI. I currently work on privacy-preserving deletion and inference-aware data management in Sharad's Lab, distributed vector search infrastructure for large-scale agentic workloads in the NetSAIL Lab, energy systems and building automation in CalPlug, and previously hyperspectral imaging and haplotype assembly with Prof. Hayes and Prof. Healy.
I'm always happy to talk about query optimization, storage engines, data deletion, agentic systems, or PhD programs. See contact below.
Papers
Inference-Aware & Privacy-Preserving Deletion in Databases
Addresses the challenge of satisfying Right-to-Be-Forgotten requests in a way that preserves differential privacy guarantees while preventing inference leakage. Introduces ILP-based deletion mechanisms evaluated across diverse UCI ML datasets, with a reproducible benchmarking framework for leakage analysis.
A Systematic Literature Review of Climate Econometrics Research
A systematic review of 300+ climate econometrics papers, evaluating methodological rigor, reproducibility, and the availability of data and code. Synthesizes trends and gaps in the field to inform future reproducible climate research.
Current & past research
Sharad's Lab
Inference-aware and privacy-preserving deletion in databases.
NetSAIL Lab
Distributed vector search infrastructure; discrete-event simulation of large-scale sharding strategies (FAISS IVF, BigANN, BatANN) for agentic AI systems.
CalPlug
Real-time occupancy-based control of lighting, plug loads, and HVAC in commercial buildings; sensor-driven automation for energy efficiency.
INsite Lab
Student data analysis across academic and health metrics; stakeholder interviews and data-driven policy recommendations for the ICS department.
Dr. Wayne Hayes Research Group
Built SAHap, a C++ haplotype assembler minimizing weighted mismatch (wMEC) across parental haplotypes via simulated annealing, extended with MaSuRCA-style super-reads and two-stage re-annealing for lower error on datasets up to 20,000 reads.
Computer Vision Lab
Quantitative analysis of hyperspectral and multispectral sensor data for environmental resilience applications (Calit2 IRT award).
Green IT Lab
Systematic literature review of 300+ climate econometrics papers, evaluating methodological rigor and reproducibility; published on SSRN.
Posters & talks
SAHap: Simulated Annealing Haplotype Assembler
Awards
Notes & posts
All posts on Medium ↗Get in touch
I'd love to hear from faculty considering PhD applicants, current grad students who can speak to a lab's day-to-day, or anyone interested in collaborating on the research above. I generally reply within a couple of days.