ACMG Variant Analyst

About the company

Deriva.ai is building the first AI-powered platform that lets clinical geneticists classify genetic variants with the intuition of an experienced scientist. By fusing fine-tuned large-language models, multi-agent orchestration, and rich biomedical data, we cut analysis time from days to minutes—giving patients faster answers and clinicians deeper insight.

About you

You live and breathe the ACMG/AMP guidelines. From PS4 to BP7, you can walk a variant through every criterion—and defend each call with iron-clad evidence. Population databases, functional studies, segregation trees, computational predictions … you pull them together into a clear, audit-ready story. You love the idea of AI automating the grunt work, but you’re just as excited to stress-test the models and make them smarter.

Responsibilities

• End-to-end ACMG classification – evaluate SNVs, indels, CNVs, and splice variants; apply ACMG/AMP (and ClinGen-specific) criteria; issue final classifications with full evidence trails.
• Evidence gathering & synthesis – mine gnomAD, TOPMed, PubMed, functional-assay papers, and family segregation data; reconcile conflicting evidence and document rationale.
• Toolset mastery – leverage VarSome, Franklin (Genoox), Ensembl VEP, Ensembl Browser, HGMD, ClinVar, and internal dashboards to cross-validate calls.
• AI feedback loop – flag edge cases, refine rule logic, and collaborate with data-science teams to improve automated scoring engines.
• Knowledge-base curation – author concise variant summaries, track re-classifications, and maintain a high-quality internal database.
• Cross-functional collaboration – partner with product, ML, and UX teams to keep outputs accurate, explainable, and clinician-friendly; contribute to user training materials.
  

Qualifications

• Clinical genetics experience – 4 + years in a diagnostic or research lab performing ACMG variant interpretation end-to-end.
• Power-user of data sources – daily comfort with VarSome, Franklin/Genoox, Ensembl VEP & Browser, gnomAD, ClinVar, HGMD, PubMed, and other population/functional databases.
• Evidence integration skills – able to weigh segregation data, de novo status, functional assays, computational scores, and literature to reach a defensible classification.
• Clear communicator – produce audit-ready reports and explain decisions to both scientists and non-experts.
• Tech fluency – experienced with bioinformatics portals or LIMS; eager to learn new SaaS tools and give rapid feedback.

Preferred extras

• Familiarity with gene- or disease-specific ClinGen expert-panel guidelines.
• Experience curating structural variants or mitochondrial genomes.
• Comfort reading basic Python notebooks or SQL queries to validate evidence pipelines.

Why Deriva.ai

• Mission with impact – every classification you issue feeds directly into patient-care decisions.
• Human + AI synergy – help shape the next generation of automated interpretation tools.
• A-player crew & good vibes – collaborate with a sharp, supportive team that teaches, learns, and celebrates wins over craft beer or a killer playlist.
• Early-stage momentum – join while the rocket is still lifting off and shape how variant interpretation evolves in the AI era.