Senior Full-Stack .NET Engineer
Systems built to hold under pressure.
Six years shipping .NET production systems — trading platforms at 99.95% uptime, LLM pipelines processing 250K+ docs a week, microservices on Azure AKS. I care about correctness, latency, and code the next engineer can maintain.
Experience
3 roles · 6 years- —Led full-stack delivery of trading and risk apps on ASP.NET Core 8 + React + Azure at 99.95% uptime.
- —Migrated a monolith to microservices on Azure AKS, cutting p95 latency ~40%.
- —Built an LLM doc-classification pipeline with OpenAI embeddings + Redis vector index — +18% accuracy at 250K+ docs/week.
- —Hardened auth with OAuth2/OIDC + JWT, closed 14 security findings, mentored 3 engineers.
- —Built clinician portals on ASP.NET Core Web API + React, used by 10K+ users daily.
- —Replaced a stored-proc layer with EF Core and tuned LINQ queries, cutting query time 62%.
- —Raised unit-test coverage from 41% to 78%.
- —Built AWS serverless workflows (Lambda + API Gateway + SQS), cutting defects ~35%.
- —Owned ASP.NET MVC/Web API features end-to-end across the payment platform.
- —Migrated classic ASP/WebForms to ASP.NET Core + Angular, halving page-load time.
- —Introduced the team's first automated CI/CD pipeline on AWS ECS.
Work
2 projectsCareerRefine
AI resume and job-application autopilot. Parses, scores, and rewrites resumes against live job descriptions using GPT-4 + embeddings. A Playwright engine auto-fills applications on Dice, Monster, and Greenhouse behind approval gates.
LLM Doc Classifier
OpenAI embeddings + Redis vector index replacing a legacy regex classification system. Processes 250K+ documents per week in production with a +18% accuracy improvement over the prior system.
Research
2 papersABQE
ATS Bias Quantification Engine
A deterministic benchmark measuring semantic sensitivity in resume–JD matching. Showed that skill phrasing and employment-gap representation cause large, statistically robust score divergence across major ATS systems.
QUANTA
Quantified Alignment & Value Transition Analysis
A model-agnostic framework measuring how LLM decision systems react to value-priority shifts. Five novel metrics — NAV, VTS, DBA, EOC, CRD — expose instability that standard fairness benchmarks miss entirely.
Stack
6 groupsLet's build something that lasts.
Open to senior IC roles and the right consulting engagements. Response time under 24 hours.