I design and deliver production-scale genomic data programs, building the infrastructure that turns raw biological data into clinical insight at consortium scale.
I own the full stack: pipeline architecture, model selection, optimization, evaluation and biological interpretation. Not just the code, the system that makes the science reproducible, scalable, and useful to partners who discover the medicine.
Common Metabolic Disease Genome Atlas · cmdga.org
$57M FNIH AMP CMD Initiative · Amgen · Eli Lilly · Novo Nordisk · Pfizer
Single-threaded technical owner and principal architect. Built end-to-end data delivery, partner onboarding and enablement infrastructure for T1D and T2D genomic research; executed single-cell ATAC-seq analyses identifying disease-associated chromatin accessibility patterns at consortium scale.
PanKbase · data.pankbase.org
$10M NIH-Funded Diabetes Research Ecosystem
Architected a comprehensive genomic + phenotypic data platform with standardized pipelines, automated QC frameworks and documentation enabling cohort-scale analysis for research partners worldwide.
Agentic Research Platform for Multi-Omics Discovery
Integrates knowledge graphs, vector embeddings, RAG pipelines and multi-agent LLM frameworks to synthesize multi-omics data for biomarker discovery and therapeutic target identification.
| Project | What | Status |
|---|---|---|
| alphagenome-explorer | Coverage tool for 714 human + 179 mouse biosamples before variant analysis | 🟢 Live |
| tert-alphageome | TERT promoter mutation analysis — chromatin, TF binding & transcription from sequence alone | 🟢 Live |
| PanKbase Data Library | Python backend and Next.js front-end for consortium data access | 🟢 Active |
Latest Writing · https://parulkudtarkar.com/blog
- TERT AlphaGenome Analysis: A Notebook Walkthrough — Methods companion: notebook setup, gnomAD control selection, AlphaGenome API calls and evaluation against published biology (Apr 2026)
- How a Single DNA Letter Change Switches On Cancer's Immortality Engine — Deep-dive into TERT promoter mutations reconstructing chromatin, TF binding and transcriptional effects from sequence alone (Apr 2026)
- AlphaGenome Coverage Explorer — Check whether your tissue or cell type is in AlphaGenome training tracks before you run (Apr 2026)
Bioinformatics & Genomics
NGS Pipelines · Single-Cell Multi-Omics (Seurat, Signac) · Spatial Omics (Scanpy, Squidpy) · ATAC-seq · RNA-seq · Variant Calling · Chromatin Accessibility Profiling · Foundation Models: AlphaGenome · Evo2 · Regulatory Network Analysis · Disease Biomarker Discovery · Target Identification
AI & Machine Learning
TensorFlow · RAG (Retrieval-Augmented Generation) · LangChain · Multi-Agent LLM Systems · Knowledge Graphs · Prompt Engineering
Cloud & DevOps
Python · AWS · Nextflow · Docker · Elasticsearch · ReactJS · TypeScript · Next.js · API Development · CI/CD
Full list on Google Scholar · 15+ publications
- FNIH AMP-CMD Leadership Meetings — Invited speaker, 2021–2025
- Intel ISEF and Chen Institute Symposium for AI Accelerated Science — Conference Jury
- Editorial Board — Database: The Journal of Biological Databases and Curation
- Peer Reviewer — AJHG · PLOS Computational Biology · Diabetes · BMC Bioinformatics · Bioinformatics · J. Endocrinology