About
Shwetha Narayanan
Applied AI Technical Lead · San Francisco Bay Area
I build production AI systems that operate in real-world environments at scale. My work spans computer vision, generative AI, and production LLM systems — from real-time fraud detection deployed across thousands of retail locations to zero-to-one GenAI products and multimodal orchestration workflows.
$300M+
annual operational savings
4,000+
retail locations scaled to
6+ yrs
production AI systems
At Walmart and Store No 8, I drove development of large-scale AI systems involving real-time inference, human-in-the-loop training pipelines, distributed model deployment, and enterprise AI workflows.
My recent work centers around evaluation-driven LLM systems using DSPy, MLflow, and automated eval pipelines to improve reliability and iterative experimentation in production AI applications.
I'm most interested in applied AI systems, AI infrastructure, evals, orchestration, observability, and building reliable AI products from zero to one.
Sunnyvale, CA
Senior Software Engineer
Walmart Inc.
Led zero-to-one image asset generation platform integrating multimodal LLM workflows and Figma MCP tooling for enterprise creative automation. Defined evaluation-driven AI optimization practices using DSPy, MLflow, and automated eval pipelines. Architected internal AI platform tooling across prompt optimization, orchestration, and multimodal generation.
Sunnyvale, CA
Software Engineer
Store No 8, Intelligent Retail Lab (Walmart)
Drove development of computer vision fraud detection for retail self-checkout, scaling from 15 → 4,000+ locations with $300M+ annual savings. Led end-to-end human-in-the-loop AI training pipelines improving detection accuracy by 9%. Architected microservices for AI model deployment, high-throughput gRPC services, and enterprise-scale data management with MLOps and CI/CD workflows.
Needham, MA
Database Operations Engineer Co-op
TripAdvisor
Built internal developer tooling to automate database schema validation and deployment workflows, improving engineering velocity.
Bangalore, India
Database Architect Intern
Envestnet Yodlee
Formulated, designed, and performed ETL processes; set up data pipelines using AWS technologies.
M.S. Computer Systems Engineering
Northeastern University
B.E. Electronics and Communication Engineering
BNM Institute of Technology