Miguel Madrid-Mencía

Hybrid: US, France

Applied ML Systems. US cloud and AI consultancy. Acquired from Qarik Group in Jan 2025 — role and scope were continuous across the acquisition.

• Designed and built multi-agent AI orchestration POCs on GCP (Gemini Enterprise Agent Platform, MCP/ADK, A2A) for a Fortune 500 pharma client, with agents connected to synthetic data for demos; proposed and designed the use-case catalog: gene-enrichment analyst, paper summarizer with iterative validation against the source paper, on-the-fly enrichment, and unified orchestration.
• Designed a configuration-file-driven Voice AI testing framework on Twilio that generates all mood and behaviour call combinations, eliminating manual test configuration and demo-affecting failures.
• Hardened CI/CD pipelines, code quality, and security standards for a major US financial services firm.
• Integrated AI agents and models into a business-document validation pipeline (Python) for the same US financial services firm, the firm’s first AI-driven document validation; sole implementer of the project’s linting within the team.
• Google Cloud Certified Partner Specialist: Gemini Enterprise Deployment + Agent Development (9 GCP skill badges); improved the firm’s internal onboarding documentation system.

Hybrid: US, France

GCP-focused cloud consultancy. Acquired by Egen.ai in Jan 2025; role and projects were continuous across the acquisition.

• Delivered a Python/SQLModel API for data ingestion into BigQuery (~15M documents), deployed on GCP with Terraform and Concourse for a leading precision medicine company, replacing a manual and semi-automated ingestion routine; corpus migration completed in under 3 months.

Hybrid: UK, France

Cloud genomics platform (Illumina Connected Analytics) and personalized medicine.

• Benchmarked privacy-preserving data-sharing approaches (homomorphic encryption vs. synthetic data) for the Illumina Connected Analytics platform.
• Built ML pipelines for genomics and GWAS on the Illumina Connected Analytics platform.
• Contributed to a pending patent in applied bioinformatics and genomic data processing — built the Hail/Python variant data-processing layer behind the ML model.

Toulouse, France

Translational cancer research center (INSERM). Network biology, epigenomics, and cancer multi-omics.

• Built reproducible Snakemake ML pipelines processing multi-omics datasets (RNA-seq, DNA methylation, 3D chromatin) across multiple cancer cohorts.
• 1 first-author and 6 co-authored peer-reviewed publications with 144+ Google Scholar citations in oncology and systems biology.
• Built GARDEN-NET, a web tool for 3D chromatin interaction analysis — published in Nucleic Acids Research.

Barcelona, Spain

European HPC center hosting MareNostrum. Life Sciences department.

• Developed a geospatial visualization platform for AQuAS to analyse health outcomes and quality indicators across Catalonia.
• Built the OpenEBench embeddable benchmarking widget (sole author) — a d3.js radial arc widget used by scientific tools across the ELIXIR European benchmarking platform.
• Primary contributor to the OpenEBench submission infrastructure, Docker sample containers, Nextflow pipeline generator, Vagrant multi-service VM, and SFTP submission service.

Madrid, Spain

Spain's national cancer research institute. Structural biology and molecular oncology.

• Developed a reproducible workflow (Structure-PPi v2.0) to identify mutation clusters at protein-protein interfaces, including web-based visualization.
• Built REST API services for the European OpenMinTeD biomedical text-mining platform, wrapping NLP tools as standardized web services.

Instituto de Salud Carlos III

Master's Thesis: Structure-PPi v2.0 — Annotating Cancer-Related Single-Nucleotide Variants at Protein-Protein Interfaces.

Universidad Complutense de Madrid

Bachelor's Thesis: Computer Vision Adapted to Optical Coherence Tomographies.

Guest Lecturer · Instituto de Salud Carlos III

Designed and delivered a 15-hour intensive module on biomedical text mining and NLP for the Master in Bioinformatics and Computational Biology (Madrid).

• Large-scale literature extraction from PubMed and PMC via the NCBI Entrez API (BioPython).
• NLP pipelines in Python/NLTK: tokenization, POS tagging, n-grams, inverted indexing, lemmatization — adapted for bioinformaticians with varied programming backgrounds.