From bioinformatics to public-interest digital tools.
This page gives the longer version: the technical background, the projects that shaped the work, and the thread connecting research, teaching, infrastructure, public knowledge, and advocacy-oriented development.
Technical summary
The short version before the longer path
My background combines biology, bioinformatics, data infrastructure, backend development, and concept design. I build digital tools that translate complex information into usable systems, especially for projects in animal advocacy, food systems, education, and open knowledge. Since 2026 I am also a certified AI manager (KI-Manager Grundlagen, TSA Bildung), covering AI strategy, prompt workflows, AI agents, process automation, and GDPR / EU AI Act compliance.
Background
Studies, research foundation, and early direction
I studied Biology at Universidad Nacional de Colombia and completed my M.Sc. in Biology at Johannes Gutenberg University Mainz. Earlier on, I also studied Architecture without completing the degree, but it strongly shaped how I think about systems, structure, and usability.
The common thread since then has been fairly consistent: working with complex information, building useful structures around it, and trying not to make technology more self-important than the people using it.
Career path
Research, infrastructure, teaching, and applied development
BioNTech · 2017–2019
Worked on internal scripts, evaluation pipelines, reports, data preprocessing workflows, and tool comparisons in the context of MHC I epitope binding prediction and neoantigen research. My Master's thesis focused on benchmarking prediction tools and evaluating indels as a source of immunogenic neo-epitopes.
DKFZ · 2019–2021
Contributed to sciReptor 2.0 and related FAIR-oriented infrastructure for storing and analyzing adaptive immune receptor repertoire single-cell data. I was also part of work on annotation standards for this type of data.
Leuphana University / DATAxHub · 2021–2023
Tested and deployed a Docker-based JupyterHub implementation before handoff to GWDG engineers, and contributed to decisions around dashboard design, authentication methods, and resource allocation for peak usage. I also taught introductory programming and data analysis with Python.
Biocyclic Vegan Network / Faba Konzepte · 2024
Supported research, course preparation, newsletter administration, and developed a basic mapping tool for certified projects worldwide. This period also fed directly into the development context of Schlachtindustrie Deutschland.
vgn.dev · 2024–today
I develop data platforms, apps, and digital tools for advocacy, education, food systems, and public knowledge projects. The work combines backend logic, data processing, interface thinking, and practical automation — increasingly including AI-supported research and workflow tooling, formalized through the KI-Manager certification (TSA Bildung, 2026).
Certifications & further training
Formal training beyond the academic path
KI-Manager (Grundlagen) · TSA Bildung, Berlin · 2026
Foundational AI management training covering AI strategy and use-case identification, prompt engineering, practical work with leading AI assistants, autonomous AI agents, business process automation with no-code tools, knowledge management, and legal compliance with GDPR and the EU AI Act. Completed with a final project.
Selected projects
A few concrete examples
Schlachtindustrie Deutschland
Backend development of a data-driven platform that maps slaughterhouse sites in Germany and makes structural harms visible: corporate networks, permits and capacities, inspection/compliance signals, environmental impacts, and documented controversies. My work spans end-to-end data acquisition (official datasets, FOI requests, curated sources), Python cleaning/normalization, structured database imports, and FastAPI delivery for a Vue-based interactive map and search experience.
Stack: Vue.js + Vite, Python (ETL with Jupyter/Pandas), FastAPI + SQLAlchemy, MariaDB/MySQL, Docker + docker-compose, and GitLab collaboration workflows (issues, merge requests, CI).
Wild Forager
Led engineering of a data-heavy mobile platform that ingests and curates open biodiversity sources, including GBIF DwC-A, Wikimedia Commons metadata, OpenStreetMap and CARTO tile ecosystems, and source-labeled records such as iNaturalist and NABU naturgucker. Built Python, SQLite, and compressed export pipelines to deliver compact occurrence datasets to a Flutter application, and implemented data contracts, bundle and manifest versioning, provenance mapping, protected-area geofencing workflows, offline cache fallbacks, and policy-aware moderation flows.
Market Gardening Veganien · 2026
Concept and end-to-end development of a mobile-first field app for vegetable growing at a sanctuary farm: live bed map with per-bed status, weather-aware day planning (Open-Meteo), crop catalogue and plant knowledge base, succession planning, and irrigation control of ESP32/ESPHome valves with on-device schedules — all synced in real time via Supabase for the whole team.
DATAxHub
A JupyterHub implementation introduced in the mandatory course "DATAx: Data analysis with Python", supporting 1,300+ first-semester students across programs with browser-based coding/data analysis. Contributed to Docker-based testing/deployment preparation and platform decisions for authentication, dashboarding, and peak resource planning.
Blog article: Introducing DATAxHub.
IsThisAnExpert · 2020
A Twitter bot concept and prototype for estimating the credibility of users posting about climate issues, combining scoring logic, expert-source aggregation, and a MySQL-backed backend.
Twitter Bots · 2021–2023
Built and maintained bots for science and advocacy communication, including a MyBrainMyChoice bot amplifying evidence-based drug policy discourse and @VgnBot, a short-lived vegan-impact initiative. Projects ended when Twitter API access constraints made continued operation impractical.
Reference: MyBrainMyChoice Twitter Bot.
Scireptor AIRR exchange tool · 2020–2021
Co-developed import/export tooling for AIRR-compliant single-cell datasets in the iReceptor Plus context, aligned with MiAIRR extension work. Built support for Common Exchange Format imports into the scireptor backend and validated workflows against AIRR standard v1.3.
Working principles
How I approach technical work
I care about digital infrastructure that is useful, careful, and maintainable. Good structure, reproducible workflows, and clear interfaces solve more than technical ornament ever will.
I use AI where it genuinely helps with research, workflow support, structuring information, and practical automation. Design and artistic work remain grounded in collaboration with real people.
Teaching & research
Education, mentoring, and publications
My background also includes teaching and mentoring in introductory programming and data analysis with Python, along with earlier research contributions in bioinformatics, immune receptor data standards, and experimental biology.
CV / résumé
PDF and external links
This page is also available as a PDF version generated automatically from the current page.