Shreyan Gupta

I'm Shreyan Gupta. I design infra systems for AI agents: memory, multi-agent orchestration, and the UX people actually touch. Recently shipped hermes and clawdown. Previously built RAG + compliance AI at Exelon.

Based in Philadelphia. Graduating June 2026. Open to full-time roles.

BOOK A CALL

░░░░░░░░░░░░░░░░░░░░

00:00

tap to see color

I care about the things most people skip. The edge cases, the impossible deadlines, the parts of a codebase nobody wants to touch. I'm currently shipping my way through building things that matter.

RESEARCH

An LLM-Based ETL Architecture for Semantic Normalization of Unstructured Data

Published in

URTC 2025

An LLM-Based ETL Architecture for Semantic Normalization of Unstructured Data

Shreyan Guptasole author

Read Paper Structura

scroll

EXPERIENCE

DREXEL UNIVERSITYB.S. Computer Science · 2022 — 2026

DREXELB.S. Computer Science2022 — 2026

DREXEL UNIVERSITY@ Undergraduate Researcher

PHILADELPHIA, PA

DEC 2025 — PRESENT

PART-TIME

DEC 2025 — PRESENT

PART-TIME

2 research roles — TeX-Base Lab, VI Lab

click to expand ↓

UNSW · GLOBAL EXCHANGEsydney, australia · sep — dec '25

HANDSHAKE@ MOVE AI Fellow

SEP 2025 — PRESENT

CONTRACT

SEP 2025 — PRESENT

CONTRACT

evaluating frontier LLMs on reasoning + accuracy + safety & writing structured r...

click to expand ↓

EXELON@ Data Scientist

PHILADELPHIA, PA

SEP 2024 — MAR 2025

CO-OP

SEP 2024 — MAR 2025

CO-OP

▪automated SPCC risk assessment across 12.5k assets. LLM structuring hit 99.4% accuracy, saved ~$1.2M/yr in regulatory overhead
▪built PySpark pipelines processing 30M+ geospatial records with H3 clustering on Azure Synapse for predictive modeling
▪designed RAG ingestion for 3.5k docs at ~35ms per file, shipped as Dockerized services on Azure with Terraform + CI/CD

ELAR INDUSTRIES@ Software Developer

MAR 2023 — SEP 2023

CO-OP

MAR 2023 — SEP 2023

CO-OP

▪built NLP-driven product search with BERT embeddings + XGBoost re-ranking. sub-100ms latency end to end
▪shipped prompt-based extraction over OCR'd PDFs at ~87% recall, cutting manual data entry by 60%+