Software for life sciences

I architect research data and build AI tooling for biotech and research teams.

I am a software engineer who speaks biology, so your scientists don’t have to translate.

I’m Bek Baymuradov. I worked on the ENCODE Data Portal at Stanford from 2015 to 2019 and have consulted independently since.

Recent clients
Sanavia Oncology, Two Bear Capital, CIQ, New York Genome Center, Columbia University
Earlier
Stanford ENCODE Data Coordination Center, Lawrence Berkeley National Laboratory, NERSC

For biotech founders and heads of computational biology

Your first pipelines, data model and AI tooling

Early on you need working pipelines, a data model you can build on and AI tools for the science, often before you have a data team.

What I take on

  • Genomics and analysis pipelines on cloud or high-performance computing
  • Data models, metadata schemas and warehouses
  • Search, summaries and dashboards over scientific data
  • AI protein-structure models applied to antibody design

Relevant work

  • Sanavia Oncology. Computational antibody design using AI protein-structure models, supporting the development of antibody-drug conjugates for cancer.
  • CIQ. Packaged AlphaFold and Boltz-2 for Fuzzball, CIQ’s high-performance computing platform, so researchers can run them with one click.
  • ENCODE, Stanford. Portable, cloud-agnostic pipelines for DNA methylation and chromatin accessibility analysis.

Tell me what you’re building and where your data lives today.

Start a conversation

For academic labs and research consortia

Portals, metadata systems and pipelines for shared data

Shared research data needs a portal people can search, consistent metadata and pipelines that others can run. I did this work for ENCODE at Stanford from 2015 to 2019.

What I take on

  • Data portals with faceted search
  • Metadata databases and schemas
  • Portable, cloud-agnostic analysis pipelines

Relevant work

  • ENCODE Data Portal. Maintained the metadata database and web backend of the public portal for ENCODE, the international follow-on to the Human Genome Project.
  • New York Genome Center and Columbia University. A blockchain platform for secure sharing of clinical and genetic data. Published in Nature Medicine, 2024.
  • ENCODE papers. Author on papers describing the ENCODE portal and its metadata storage system.

Tell me about your data and the people who need to use it.

Start a conversation

For investors

Independent technology assessment for biotech and techbio

I review a company’s data, pipelines, models and software as an engineer who has built the same kinds of systems. You get an independent technical read in plain language.

What I review

  • Data architecture and infrastructure
  • Genomics pipelines and computing platforms
  • AI and machine learning models for drug discovery

Relevant work

  • Two Bear Capital. Technology assessment.
  • Hands-on range. Genomics data portals, analysis pipelines, supercomputer service tracking at NERSC, AI protein-structure models and clinical data sharing.
  • Virtual Cell Challenge 2026. Competed solo, building machine learning models that predict how human cell lines respond to gene knockdowns.

Tell me about the company and your timeline.

Start a conversation

Track record

Independent since 2019. Before that, Stanford and Lawrence Berkeley National Laboratory.

Consulting clients

Sanavia Oncology

Computational antibody design using AI protein-structure models, supporting the development of antibody-drug conjugates for cancer.

Two Bear Capital

Technology assessment.

CIQ

Packaged scientific pipelines, including AlphaFold and Boltz-2, for the workflow catalog of Fuzzball, CIQ’s high-performance computing platform, so researchers can run them with one click.

New York Genome Center and Columbia University

Co-developed a blockchain platform for secure sharing of clinical and genetic data. Published in Nature Medicine, 2024.

Read the paper

Before Ocean Source

Stanford University School of Medicine

to

Role
Software Developer, ENCODE Data Coordination Center

ENCODE, the Encyclopedia of DNA Elements, is the international follow-on to the Human Genome Project. I worked on the team that runs its public data portal, used by researchers worldwide.

  • Maintained the metadata database and web backend of the ENCODE Data Portal.
  • Built portable, cloud-agnostic pipelines for DNA methylation and chromatin accessibility analysis.
  • Designed on-demand Elasticsearch deployment on AWS and faceted search of genomic regions across sequence data files.
  • Led a team in the ENCODE Imputation Challenge.
  • Mentored students in Stanford’s Genomics Research Internship Program.

Lawrence Berkeley National Laboratory, NERSC

to

Role
Software Engineer
  • Built software to track outages and service status of NERSC’s supercomputers.
  • Contributed to the system that tracks users’ allocated computing quotas.

Independent project

Virtual Cell Challenge 2026, Arc Institute

Competed solo. Built machine learning models for zero-shot prediction of how human cell lines respond to CRISPR gene knockdowns, in a global competition with more than 1,500 teams on the leaderboard.

Publications

I am an author on nine publications. On the Nature papers and the 2026 preprint, I am credited as a member of the ENCODE Project Consortium.

  1. 2026
    The Encyclopedia of DNA Elements

    bioRxiv preprint ENCODE Project Consortium

    doi.org/10.64898/2026.07.06.731365

  2. 2024
    A framework for sharing of clinical and genetic data for precision medicine applications

    Nature Medicine Second author

    doi.org/10.1038/s41591-024-03239-5

  3. 2020
    Expanded encyclopaedias of DNA elements in the human and mouse genomes

    Nature ENCODE Project Consortium

    doi.org/10.1038/s41586-020-2493-4

  4. 2020
    Perspectives on ENCODE

    Nature ENCODE Project Consortium

    doi.org/10.1038/s41586-020-2449-8

  5. 2020
    New developments on the Encyclopedia of DNA Elements (ENCODE) data portal

    Nucleic Acids Research

    doi.org/10.1093/nar/gkz1062

  6. 2019
    The ENCODE Portal as an Epigenomics Resource

    Current Protocols in Bioinformatics

    doi.org/10.1002/cpbi.89

  7. 2018
    Prevention of data duplication for high throughput sequencing repositories

    Database (Oxford)

    doi.org/10.1093/database/bay008

  8. 2018
    The Encyclopedia of DNA elements (ENCODE): data portal update

    Nucleic Acids Research

    doi.org/10.1093/nar/gkx1081

  9. 2017

About

I’m Ulugbek (Bek) Baymuradov, founder of Ocean Source. I started it in 2019, when I left Stanford’s ENCODE Data Coordination Center to consult independently. Before Stanford I was a software engineer at NERSC, Lawrence Berkeley National Laboratory. I hold a BA in Economics from Stanford.

I work remotely with US clients.

Tell me about your project

The form asks for your name, your email and a few lines about the work.

You can also message me on LinkedIn.