About me

From the lab bench to the command line

I’m Salvatore, based in London. I moved into bioinformatics from the lab bench. I spent two years preparing NGS libraries for clinical preimplantation genetic testing at CooperGenomics, then four years in the UCLH stem cell laboratory supporting clinical and ATMP trials, where I also built Python tools to replace manual data reconciliation. That background shapes how I analyse data: I know how sequencing data is generated, where it fails QC, and what audit-ready documentation looks like.

For my MSc in Bioinformatics at Atlantic Technological University (2025–2026), my dissertation prioritised GPNMB as a Parkinson’s disease-relevant protein using Mendelian randomisation and colocalisation, and I followed it up with a donor-aware single-nucleus RNA-seq re-analysis. I care about honest inference: donors rather than nuclei as replicates, sensitivity analyses that can weaken a result, and conclusions no stronger than the evidence.

Curriculum Vitae LinkedIn GitHub Email
Based in
London, UK · remote-friendly
Right to work
United Kingdom & European Union
Languages
ItalianNative
EnglishFluent
PortugueseFluent
SpanishAdvanced

Experience

2018 – 2026

Scientific AI evaluation

Jan 2026 – presentContract · remote

Mercor

Bioinformatician · Scientific AI Evaluation

Designed scientific-reasoning evaluations that test whether frontier language models can infer hidden biological and computational rules through controlled experimentation.

  • Authored 106 accepted tasks (Jan–Jul 2026) for a frontier model’s bioinformatics benchmark. Each is a hidden Python function implementing a cited genomics or biostatistics rule, with unit tests, that the model must reverse-engineer by choosing inputs and reading outputs.
  • Calibrated difficulty from model trajectory data using ablation ladders and adversarial foils: textbook solutions scored below 50% while the full solution reached 100%.
  • 80% first-pass approval through peer and client review.
  • Expert evaluation of AI agent runs on preclinical drug R&D tasks, and blinded A/B comparisons of coding models on large open-source codebases.
  • Scientific AI
  • Python
  • Unit Testing
  • Biology
  • Benchmark Design

Earlier: shorter AI training and evaluation contracts with Outlier and Micro1, covering biology answer review, adversarial multimodal prompts and reasoning data.

Clinical laboratories

  1. Sep 2021 – Dec 2025London, UK

    UCL Hospitals NHS Trust

    Specialist Biomedical Scientist · Stem Cell Laboratory

    • Built Python/Excel tracking tools to replace manual inventory reconciliation, reducing inventory errors by ~30% and saving ~10 hours a week.
    • Managed traceability and audit-ready records for 11 clinical and ATMP trials, including CAR-T therapies.
    • Led the digitisation of SOPs and QA documentation, standardising data handling across workflows.
    • Processed and cryopreserved PBSC, bone marrow, DLI and CD34+ products, and generated CD3+/CD34+ flow cytometry data for time-critical clinical decisions.
  2. Jul 2019 – Sep 2021London, UK

    CooperGenomics

    Laboratory Scientist · Clinical Genomics

    • Processed embryo samples for PGT-A, PGT-SR, and PGT-M testing within a high-throughput clinical genomics pipeline.
    • Prepared and QC’d NGS libraries at 96–192 samples per run (500+ samples a week) to clinical turnaround targets.
    • Programmed and validated Mosquito HV and Dragonfly liquid handlers, improving workflow scalability and reproducibility.
    • Contributed to SOP writing and review, strengthening ISO-compliant laboratory practice.
  3. Dec 2018 – Jun 2019Leicester, UK

    Leicester Royal Infirmary

    Biomedical Laboratory Assistant · Cytology

    • Managed sample reception and prepared specimens for Papanicolaou staining.
    • Maintained reagents and ensured sample integrity end-to-end.

Toolkit

What I use

Programming
  • Python
  • R
  • SQL
  • Bash
Workflows
  • Nextflow (DSL2)
  • Docker
  • Galaxy
  • Conda
  • Git
  • Linux
NGS / Omics
  • RNA-seq
  • WGS
  • fastp
  • Cutadapt
  • STAR
  • Salmon
  • featureCounts
  • BWA-MEM2
  • samtools
  • Picard
  • bcftools
  • SnpEff
  • mosdepth
  • MultiQC
  • DESeq2
  • IGV
Single-cell
  • Scanpy
  • AnnData
  • Pseudobulk DE
  • PyDESeq2
  • speckle / propeller
  • Leiden clustering
Statistical genetics
  • Mendelian randomisation
  • TwoSampleMR
  • Colocalisation
  • coloc
  • SuSiE
  • susieR
  • GWAS
  • pQTL / eQTL
  • LD clumping
Statistics
  • Survival analysis (Cox, Kaplan–Meier)
  • Generalised linear models
  • Multiple testing (FDR)
  • Meta-analysis (IVW, Cochran’s Q)
  • Robust MR estimators (MR-Egger, weighted median)
  • Bayesian inference
  • Sensitivity analysis
  • Donor-level replication
  • Cross-validation
  • AUROC
Machine learning
  • scikit-learn
  • XGBoost
  • PyTorch
Scientific AI evaluation
  • Blackbox tasks
  • Deterministic Python functions
  • Unit-test design
  • Prompt writing
  • Scientific reasoning benchmarks

Education

3 degrees · 2014 – 2026

  1. 2025 – 2026

    MSc, Bioinformatics

    Atlantic Technological University · Letterkenny, Ireland · Remote

    DissertationPrioritising GPNMB as a Parkinson’s disease-relevant protein using Mendelian randomisation and colocalisation · case study

  2. 2021 – 2023

    MSc, Cell & Gene Therapy

    University College London · London, UK

    DissertationExpansion and Preservation of Haematopoietic Potential in Human Amniotic Fluid Stem Cells for Therapeutic Applications

  3. 2014 – 2017

    BSc, Biomedical Science

    University of Catania · Catania, Italy

    DissertationCytotoxicity assays using SIRC, ARPE-19, and HRPE cells

Certifications

Bioinformatics, ML/AI and data engineering first; more under Additional

Johns Hopkins University
Genomic Data Science Specialization
Wellcome
Bioinformatics for Biologists: Linux, Bash, R · Analysing Genomics Datasets
DE<code>LIFE
Genomes, Networks & Pathways · Data Science & Machine Learning
Ulster University
Coding Skills for Biologists
Coursera
Access Bioinformatics Databases with Biopython
IBM
Machine Learning · AI Engineering · Data Engineering
AdditionalData analytics · BI · Cloud
Google
Data Analytics · Advanced Data Analytics · Business Intelligence · IT Automation with Python · Project Management
AWS
Cloud Practitioner Essentials · Cloud Solutions Architect
Google Cloud
Architecting with Google Kubernetes Engine
SAS
SAS Programming 1: Essentials · SAS Programming 2: Data Manipulation Techniques
freeCodeCamp
Data Analysis with Python · Relational Databases · Scientific Computing
Le Wagon
Data Visualization with Tableau

Get in touch

I’m looking for bioinformatics roles in statistical genetics, transcriptomics and NGS analysis, especially where clinical or cell and gene therapy experience helps. Based in London, open to hybrid and remote.

barbagallo1@hotmail.it
Curriculum Vitae LinkedIn GitHub