Rajat Aggarwal

About Rajat Aggarwal

I’m a Senior Engineering Manager with 13+ years building large-scale data, API, and AI/GenAI platforms. I design and deliver systems that make engineering teams productive, enable AI-first automation, and ensure data reliability at scale. My work spans platform architecture, LLM/GenAI integration, backend services, and observability.Key strengths:Architecting GenAI & LLM systems: agentic workflows, RAG, semantic indexing, prompt engineering, grounding LLMs to data.Building metrics & semantic platforms: metric definitions, governance, lineage, aggregation, multi-engine compute.Designing distributed backend systems & APIs: REST/gRPC microservices, FastAPI, Spring Boot, backend-for-frontend.Automating data pipelines & SQL workflows: AI-assisted SQL transpilation/optimization, incremental/backfill strategies.Ensuring platform reliability: regression testing, change impact analysis, observability (Datadog/ELK/Splunk).Cloud & infra: Kubernetes, EKS, Terraform, AWS/GCP/Azure, containerization, CI/CD.I lead cross-functional teams (15–100+), collaborate with research and product groups, and deliver measurable platform impact — cost savings, improved velocity, and stronger data trust. I enjoy mentoring engineers, simplifying complex systems, and shipping practical AI features that solve real problems.Search keywords / Skills: GenAI, LLM, RAG, MCP Server, semantic layer, SQL transpilation, SQLGlot, StarRocks, Presto, Spark, BigQuery, REST API, gRPC, FastAPI, Spring Boot, Python, Java, Scala, Spark, Kafka, Airflow, Kubernetes, EKS, Docker, Terraform, Datadog, ELK, Splunk, CI/CD, data mesh, MDM, metrics platform, data governance, observability, regression testing, backfill, incremental load, agentic AI, prompt engineering.

Experience

1. Senior Engineering Manager

   Expedia Group

   Oct 2022 — Present · Gurugram, IN

   Led the platform team building a global metrics & AI platform that enforced consistency, governance, and self-serve analytics across distributed systems, driving multi-million dollar operational efficiencies.Architected and shipped GenAI-enabled workflows: LLM-backed SQL optimization, AI-assisted SQL transpilation across engines, and NLQ-based dataset onboarding.Designed and operated distributed backend services & APIs (REST/gRPC/FastAPI) for onboarding, metadata serving, SQL compilation, and agent orchestration.Built a Generic MCP server to support agentic workflows for onboarding automation, schema validation, automated documentation, and prompt-driven data assistance.Scaled platform to support metrics with automated backfills, incremental pipelines, engine-aware SQL generation, and compute orchestration for heavy operations.Implemented AI-powered regression previews and automated change-impact analysis to surface downstream consequences of metric or SQL changes before release.Introduced a semantic proxy / metadata API layer that provides lineage, explainability, and contextual prompts to LLMs, improving trust and traceability.Championed adoption of high-performance engines and caching strategies (e.g, StarRocks) for virtual analytics and fast interactive queries.Grew and mentored a cross-functional team (15+) focused on platform reliability, automation, and integrative AI features.