Tejvir Virk — Full Stack Engineer
I’m Tejvir Virk, a full stack engineer in Dublin, CA. I spent 3 years at AWS building scalable backend services, REST APIs, web applications, and LLM powered tools for enterprise, healthcare, and government clients. I turn ambiguous requirements into shipped software, then iterate on direct customer feedback.
A foundation in
computer science.
University of California, Santa Cruz
B.S. in Computer Science, September 2019 to June 2023.
Amazon Web Services
Full Stack Engineer (Delivery Consultant)
Three years building scalable backend services, REST APIs, and web applications for enterprise, healthcare, and government clients. Weighed design trade-offs to turn ambiguous requirements into shipped software, with frequent follow-on releases driven by production usage and direct customer feedback.
LLM powered physician call documentation.
Built and shipped a HIPAA compliant app for Pfizer sales reps using React, Python on AWS Lambda, and Claude. Summarizes, tags, and redacts PHI from physician calls.
90% REDUCTION IN DOCUMENTATION TIMEFeatures and APIs for lab operations.
Built features and REST APIs for a lab management platform using React, Python/FastAPI, PostgreSQL, and Docker. Covered patient intake, lab orders and updates, and multi-stage workflows.
1,000+ SCIENTISTS USING THE PLATFORM DAILYMore accessible government documents.
Extended a Python Section 508 tagging tool at CMS to handle PDFs with tables of contents, malformed lists, and paragraphs spanning pages.
30 MIN → UNDER 1 MIN MANUAL TAGGING PER FILETested, reviewed, and released.
Wrote automated tests and technical documentation for React frontends and Python backends, reviewed code, and shipped releases through Git based CI/CD.
Security Specialty 2024
Certified Developer Associate 2025
Solutions Architect Associate 2026
AI Deployment Auditor
From a public repo to a reviewable audit.
Architected and shipped solo with a React frontend and FastAPI backend. Clones any public GitHub repository and runs an agentic RAG pipeline against OWASP, NIST, and AWS policy documents, retrieving context through Pinecone vector search across 12 automated checks.
Implemented an LLM as judge evaluation layer using GPT-4o mini that scores each check 0 to 10 with policy citations, keeping results reproducible and reviewable.
Shipped a custom document upload feature so users can inject their own compliance standards through the same ingestion pipeline, extending the tool to new domains without code changes.
Explore the live project ↗Skills
- LANGUAGES
- Python · TypeScript · JavaScript · SQL
- FRAMEWORKS
- FastAPI · Pydantic · React · REST APIs
- DATABASES
- PostgreSQL (relational) · DynamoDB (NoSQL) · Pinecone (vector)
- CLOUD & DEVOPS
- AWS (Lambda, S3, EC2, CloudFormation, CDK) · Docker · Git · CI/CD · Linux
- AI / ML
- Claude API · Claude Code · Amazon Bedrock · LangGraph · RAG · LLM as judge · Evaluation driven development
What should we
build next?
Have an interesting problem to solve? Let’s start with a conversation.
tejvirv@gmail.com ↗