AI DATA SCIENTIST
FOR HEALTHCARE AND PHARMACEUTICAL INDUSTRYI am Corina Roca. Software developer with Bioinformatics background and + 8 years of experience in the pharmaceutical industry.
SERVICES
Services That I Provide
Full-stack development
Locally or in the cloud. I work with JS, Python, R and SQL or non-sql databases to organize and bring your data from raw to customer. From monolitic architectues to containerized software.
Learn MoreData Science ad-hoc projects
With a focus in healthcare and pharmaceutical industry needs, I collaborate/support data science teams and epidemiologist to bring the most of the value from the data to enable medical discoveries.
Learn MoreAI Tech Stack Consultancy
Providing guidance about newest AI tools and technologies that could help your business to automate/simplify pipelines.
Learn MoreWORK
My Recent Projects
Data Science | RWE insights for ES-SCLC
2024 New Publication under review to showcase insights after statistical analysis in RWE set.
See our previous publication (2023)
WebApp | Clinical trials reporting with R and Teal
R shiny dashboards using Teal packages (v.2024) or others from Pharmaverse
See Here
AI Agents and Agentic Workflows (2025)
Local or in cloud Agentic Workflows to integrate with structured and non-structured data. Model refinement and Evaluations in place to stablish guardrails and certain levels of accuracy.
DEMO of the 🏆 Winner project at OT Hackathon '25
Data curation/Harmonization (2016-2017)
My background in pharmacology and bioinformatics allows me to perform some data curation over certain datasets. Currently is useful to me when creating "evals" for Scientific Agentic workflows.
Example of the model of the month (previous work at EBI)EVENTS & COLLABS
Most recent attended workshops and hackathons:
Federated Learning FlowerAI Collaborative Hackathon
"Creating a Agent to run Federated learning for macular degeneration model across Africa and UK sites" project.
Using the Federated Learning FlowerAI app to bring data across two countries to train a master model. Flower-Seed app
Fourth International Biohackathon (2026) Cambridge-Ant
2 projects: Nexons visualizer (for OxfordNanopore) and LLM evaluation for pipeline construction.
Project 1: Improve the current Nexons pipeline with a minimap2 flow and Nexons Visualizer. Also porting the tool to other long-read sequencing analyzers. Nexons viewer
Project 2: LLM evaluation to build a DQ pipeline. Several "vibe check" teams unified their approach to evaluating the output and classifying whether the LLM performed well or not on the same task.
Claude Code Cambridge monthly meetings
Monthly talks to see how Claude AI is used and the degree of success in several big and small companies.
Talks provided by AstraZeneca, Redgate and more (posting details soon)
Software Crafters - 11th August 2026
"Trusting the Untrusted: Why Domain-Specific LLM Eval Matters" presentation
A 5-minute talk about the current problem and approaches to the Observability and Evaluation of LLM agents in biological science. Other activities are dojo software architecture (collaborative architectural software problems to solve). And Dojo Code.
Experience
My Education + Work History
Education
eLfH NHS courses (2026)
Applying a Health Psychology approach to Workforce Redesign (HPA)
Special focus in the topic: Digital and management skills (Essentials of digital clinical safety & Data Security Awareness.)
FlowerAI Federated Learning Workshop (2026)
Federated learning certification using Flower AI
How to train a master/common machine learning model from different simulated local sites without sharing the raw data. This workshop provide a hands-on example and code to unify ML parameters across different simulated clinical sites, withouth sharing the raw data, to train a common ML in-real time. Keeping privacy and security first. FlowerAI framework, Jupyter notebook and Python language
CDISC course (2024)
SDTM and JSON-dataset
Introduction to SDTM format and JSON-dataset structure to store clinical trial data
MSc Computational Biology(2014-2016)
Instituto Carlos III (UCM)
Bioinformatic tools, analysis and learning about multi-omics data and computational approaches to get insights for drug discovery (Gene Enrichment analysis, Protein structural homology, Machine Learning, ). MSc thesis. Thesis within a biotech industry with focus in diagnosis project linking cancer blood clinical samples with drug resistance predictions.
Pharmaceutical industry degree (2008-2014)
Universidad de Alcala + 1 Year Erasmus at University of Warsaw
Deep knowledge in Biochemistry, Pharmakokinetics, Pharmakodynamics, Biology, Statistics and Pharmaceutical technology. Final laboratory year within the Immunology research laboratory in the Medical university of Warsaw. Topic: Macrophage behaviour when Ligand receptor interaction of the protein CD220
Work Experience
Freelance Software developer for research
FairCurator Ltd (2026-Currently)
Mantaining and developing software for drug discovery research. Participating in conferences with focus in infrastructure for AI in biomedical research.
AI software engineer
Astex Pharmaceuticals (2024-2026)
Mantaining and developing scientific web platform for target discovery. Developing AI stack technnology in-house.
RWE Data scientist Consultant
AstraZeneca (2022-2023)
Data architect - Consultant
Roche (2018-2022)
Data Governance team. Full stack developer, creating apps upon request to address specific company issues in a more user-friendly way.
EBI-EMBL
Data curator (2016-2017)
Curator of cuantitative and qualitative databases (Biomodels and Reactome). Also Researcher for an auto-immune orphan disease, using bioinformatic tools. EBIEMBL Heilderberg conference 2016