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

I’m an AI Engineer with 5+ years of experience building production-grade AI and data systems across healthcare, semiconductor, and enterprise environments. I specialize in agentic AI, RAG, and LLM orchestration, with hands-on delivery of real-world workflows and scalable backend AI services. I’ve led the transition of agentic systems from internal prototypes to production, improved clinical decisioning performance through grounded policy-backed retrieval, and built large-scale ML pipelines and data quality safeguards. I enjoy turning complex data and model workflows into reliable, observable, maintainable systems that teams can confidently use.…I’m an AI Engineer with 5+ years of experience building production-grade AI and data systems across healthcare, semiconductor, and enterprise environments. I specialize in agentic AI, RAG, and LLM orchestration, with hands-on delivery of real-world workflows and scalable backend AI services. I’ve led the transition of agentic systems from internal prototypes to production, improved clinical decisioning performance through grounded policy-backed retrieval, and built large-scale ML pipelines and data quality safeguards. I enjoy turning complex data and model workflows into reliable, observable, maintainable systems that teams can confidently use.

Shree Lavalika

AI Engineer, Data Scientist, Software Engineer, +1





I’m an AI Engineer with 5+ years of experience building production-grade AI and data systems across healthcare, semiconductor, and enterprise environments. I specialize in agentic AI, RAG, and LLM orchestration, with hands-on delivery of real-world workflows and scalable backend AI services. I’ve led the transition of agentic systems from internal prototypes to production, improved clinical decisioning performance through grounded policy-backed retrieval, and built large-scale ML pipelines and data quality safeguards. I enjoy turning complex data and model workflows into reliable, observable, maintainable systems that teams can confidently use.…I’m an AI Engineer with 5+ years of experience building production-grade AI and data systems across healthcare, semiconductor, and enterprise environments. I specialize in agentic AI, RAG, and LLM orchestration, with hands-on delivery of real-world workflows and scalable backend AI services. I’ve led the transition of agentic systems from internal prototypes to production, improved clinical decisioning performance through grounded policy-backed retrieval, and built large-scale ML pipelines and data quality safeguards. I enjoy turning complex data and model workflows into reliable, observable, maintainable systems that teams can confidently use.

Available to hire

I’m an AI Engineer with 5+ years of experience building production-grade AI and data systems across healthcare, semiconductor, and enterprise environments. I specialize in agentic AI, RAG, and LLM orchestration, with hands-on delivery of real-world workflows and scalable backend AI services.

I’ve led the transition of agentic systems from internal prototypes to production, improved clinical decisioning performance through grounded policy-backed retrieval, and built large-scale ML pipelines and data quality safeguards. I enjoy turning complex data and model workflows into reliable, observable, maintainable systems that teams can confidently use.

Skills

Experience Level

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Intermediate

Intermediate

Intermediate

Language

Work Experience

Software Engineer, Agentic AI at Optum

June 1, 2024 - Present

Automated end-to-end prior authorization decisions, scaling capacity to handle 1,000–5,000 daily requests without increasing headcount. Engineered RAG pipelines pulling from live clinical policies and medical evidence to provide grounded, policy-backed recommendations. Reduced manual review volume by 25% using confidence-based routing that automatically resolved 1 in 4 borderline cases, and decreased authorization response times by 30% by optimizing vector retrieval, restructuring prompts, and parallelizing API calls. Lowered agent failure rates by 25% with tracing, schema validation, Redis caching, and Prometheus/Grafana observability to detect pipeline errors before production. Built reusable Python and TypeScript REST services enabling clinical ops and backend teams to integrate seamlessly, and led the transition from internal prototype to production across product, platform, and clinical ops teams.

Software Engineer, Data Science at Qualcomm

March 1, 2021 - July 31, 2022

Processed 5M+ Snapdragon telemetry records to build a daily visibility dashboard, replacing manual extraction with automated performance anomaly tracking. Reduced downstream debugging effort by deploying tabular anomaly detection models to catch CPU load, thermal, and power draw defects early in validation. Detected hidden chipset performance degradation by building sequence-aware recurrent models over time-series trace data to capture patterns missed by point-in-time checks. Improved overall reliability by implementing automated validation checks that blocked malformed telemetry records at ingestion instead of propagating failures into model outputs.

Software Engineer, Data at Red Hat

January 1, 2019 - February 28, 2021

Automated build health tracking across 5–10 Jenkins pipelines into a centralized, queryable PostgreSQL database, eliminating manual health checks. Enabled data-driven leadership decisions via schema-backed views visualizing release cadence and historical build health trends. Reduced staging-to-production discrepancies by migrating legacy pipeline jobs to Docker and OpenShift containerized environments. Saved ~300 hours annually by automating recurring data extraction and aggregation for management reporting, and improved pipeline reliability by adding upstream checks that caught and resolved data issues before silent downstream report failures.