Looks like you have JavaScript disabled. For the full Twine experience, you will need to re-enable it.

I am an AI-focused Python backend engineer with around five years shaping healthcare and insurance platforms. I build ML-backed APIs, RAG workflows, and data pipelines that convert raw clinical, claims, and policy data into reliable inference endpoints. I collaborate with product, data science, and operations teams to deploy robust systems on AWS, maintain CI/CD flows, and monitor performance and drift to ensure predictable latency during high-volume claim cycles.…I am an AI-focused Python backend engineer with around five years shaping healthcare and insurance platforms. I build ML-backed APIs, RAG workflows, and data pipelines that convert raw clinical, claims, and policy data into reliable inference endpoints. I collaborate with product, data science, and operations teams to deploy robust systems on AWS, maintain CI/CD flows, and monitor performance and drift to ensure predictable latency during high-volume claim cycles.

Vaidik Vyas

Back-End Developer, Web Developer, Full Stack Developer, +2





I am an AI-focused Python backend engineer with around five years shaping healthcare and insurance platforms. I build ML-backed APIs, RAG workflows, and data pipelines that convert raw clinical, claims, and policy data into reliable inference endpoints. I collaborate with product, data science, and operations teams to deploy robust systems on AWS, maintain CI/CD flows, and monitor performance and drift to ensure predictable latency during high-volume claim cycles.…I am an AI-focused Python backend engineer with around five years shaping healthcare and insurance platforms. I build ML-backed APIs, RAG workflows, and data pipelines that convert raw clinical, claims, and policy data into reliable inference endpoints. I collaborate with product, data science, and operations teams to deploy robust systems on AWS, maintain CI/CD flows, and monitor performance and drift to ensure predictable latency during high-volume claim cycles.

Available to hire

I am an AI-focused Python backend engineer with around five years shaping healthcare and insurance platforms. I build ML-backed APIs, RAG workflows, and data pipelines that convert raw clinical, claims, and policy data into reliable inference endpoints.

I collaborate with product, data science, and operations teams to deploy robust systems on AWS, maintain CI/CD flows, and monitor performance and drift to ensure predictable latency during high-volume claim cycles.

Skills

Experience Level

Expert

Expert

Expert

Expert

Expert

Language

English

Fluent

Work Experience

AI Backend Engineer at MetLife

August 1, 2024 - November 25, 2025

Led a RAG-based decision-support module using LangChain + FAISS to pull policy clauses and historical notes, reducing adjuster lookup time by 40%. Exposed AI summarization and scoring through a lightweight FastAPI layer for plug-and-play integration with underwriting and claims systems. Integrated Hugging Face Transformers for policy-text summarization, delivering concise snippets that reduced internal review cycles by 25%. Containerized AI components with Docker and deployed on AWS ECS with Terraform, adding autoscaling to stabilize latency during monthly spikes. Built a structured data-prep pipeline in Airflow, SQL, and Pandas to clean eligibility and adjudication records for embedding workflows, reducing refresh time from 3 hours to under 50 minutes. Implemented end-to-end CI/CD with GitHub Actions, automating model packaging, API builds, and blue-green rollouts for near-zero downtime. Added Prometheus-based observability with Python probes, tracking embedding drift and latency to r

Software Development Engineer at Vivma Software Inc.

July 1, 2023 - July 1, 2023

Built patient-risk scoring APIs using Flask and XGBoost to classify readmission likelihood from claims and vitals data, improving early-flag detection by 17%. Created a treatment-path recommendation module with Pandas and scikit-learn to compare historical outcomes and shorten clinical review time by 28%. Designed a claims-cleaning pipeline with SQL and AWS Lambda to auto-fix mapping gaps and duplicates, reducing manual adjudication effort by 40%. Built a physician-notes parser with spaCy to extract ICD/NDC terms and feed triage screens, increasing coding precision by 22%. Implemented front-end review dashboards in React with API hooks to backend scoring services, helping reviewers close high-priority cases faster. Established CI/CD with Jenkins and GitHub Actions to run unit tests, security checks, and auto-deploy scoring services, cutting release failures by 35%. Containerized ML inference and ETL jobs with AWS ECS and S3, maintaining monthly data-prep latency under 45 minutes during