Open to opportunities

Company Yardie

@companyyardie

Message

Ph.D. computational biologist designing anonymized, reasoning-focused AI benchmarks from real omics data.

United States

Message

What I'm looking for

I seek roles designing and validating anonymized, reasoning-focused AI benchmarks or bioinformatics pipelines, collaborating with ML teams on challenging, multi-omics tasks.

I am a Ph.D.-trained computational biologist with 7+ years converting messy real-world omics into structured, model-ready evaluation challenges that measure reasoning rather than retrieval. I design high-complexity "mystery problems" with biologically grounded single definitive answers and 20–40 carefully engineered distractors.

I build fully reproducible analysis pipelines using Python, R, bash and integrate tools such as samtools, bcftools, STAR, HISAT2, BLAST, DESeq2 and GATK, and optimize workflows with Jupyter, Conda, Docker and cloud environments (AWS/GCP). I validate benchmark integrity through independent analytic paths, orthogonal pipelines, and robust QC to ensure datasets reflect authentic biological signals.

I have extensive freelance consulting experience (Upwork, Fiverr, Toptal) and institutional work designing multi-omics benchmarking, metadata-stripping and leakage-prevention strategies, and scoring rubrics used to evaluate LLM reasoning and multi-step biological interpretation.

Experience

Work history, roles, and key accomplishments

Current

Senior Computational Biologist

Current

Freelance Consultant

Jan 2020 - Present (5 years 11 months)

Designed and delivered 75+ reasoning-focused AI benchmark tasks and fully anonymized biological datasets from GEO/SRA/ENCODE, achieving Top Rated Plus status on Upwork and ensuring robust validation and metadata-masking to prevent dataset leakage.

Python R DESeq2 SAMtools GATK Docker AWS Anonymization Benchmark Design

Computational Biology Specialist

Genomics Research Institute

Jan 2020 - Dec 2023 (3 years 11 months)

Led multi-omics benchmarking projects and authored domain-specific question templates to evaluate model reasoning on PCA/UMAP and DEG outputs, implementing algorithmic distractor construction and orthogonal verification of solutions.

DESeq2 GATK Noise Modeling Benchmarking

Bioinformatics Analyst

University of Florida Health

Jan 2016 - Jan 2020 (4 years)

Managed large-scale WGS/WES/RNA-seq projects, built reproducible pipelines for preprocessing, batch-effect correction, and gene-level summarization, and produced QC-ready matrices and visualizations for downstream analysis.