Yogesh Watamwaar

Led technical design and development of Java and Spring Boot microservices for internal AI tools, moving prototypes into secure, reliable production services. Built and operated Java, Node.js, and Python REST services end-to-end (design through deployment, monitoring, and on-call) handling millions of daily requests. Developed React and TypeScript frontends and created shared component libraries adopted by multiple product teams, improving UI development speed. Implemented conversational AI capabilities using LLM APIs, RAG, and vector search; automated document ingestion and retrieval pipelines and improved answer-grounding accuracy while reducing hallucinations. Optimized Elasticsearch indexing/search across millions of documents; implemented CI/CD with GitLab CI, GitHub Actions, and Argo CD with automated testing and progressive deployments. Improved API contract quality, schema validation, structured error handling, and documentation; strengthened observability with Dynatrace/Promet

Led technical design and development of Java 17 and Spring Boot microservices for internal AI tools, taking prototypes into secure, reliable production services. Built and operated Java, Node.js, and Python REST services from system design through deployment, monitoring, and on-call support for systems handling millions of daily requests. Developed React and TypeScript frontends for high-throughput APIs; created shared component libraries adopted by four product teams, reducing UI development time by 35%. Designed conversational AI capabilities using LLM APIs, retrieval-augmented generation (RAG), and vector search, automating employee support and deflecting 42% of routine IT tickets. Built document-ingestion and retrieval pipelines using chunking, embeddings, and hybrid relevance scoring; achieved 93% answer-grounding accuracy and reduced hallucinated responses by 60%. Optimized Elasticsearch indexes, queries, and aggregations across 12 million documents, reducing average search latency from 850 ms to 190 ms. Implemented CI/CD pipelines with GitLab CI, GitHub Actions, and Argo CD, automating tests, image builds, and progressive deployments; increased release frequency 3x. Improved API contracts, schema validation, structured error handling, and documentation, reducing frontend/backend integration defects by 25%. Established structured logging, distributed tracing, and monitoring with Dynatrace, Prometheus, and Grafana; reduced mean time to detect production issues by 45%. Implemented SSO, OAuth, RBAC, secrets management, and audit logging; created usage and LLM-cost dashboards that supported a 28% reduction in annual AI spend. Used Claude, Cursor, and GitHub Copilot for implementation and test generation, increasing feature throughput by 30% while retaining code review and quality ownership. Mentored four engineers and led design reviews for cross-team initiatives, standardizing async messaging patterns with Kafka and SQS that cut duplicate processing incidents by 50%.

Owned backend product and platform development from design through production for high-volume mortgage transaction processing and regulated financial workflows. Architected distributed Python, Go, and Ruby on Rails services supporting large-scale financial transactions; improved throughput by 40% while maintaining sub-second API response times. Developed REST APIs with Node.js, Python, and Go, enabling validated data exchange across frontend applications, backend services, and downstream financial systems. Designed a PostgreSQL and Redis risk-scoring engine evaluating more than 30 behavioral signals per request; enabled analyst-managed thresholds and reduced repeat abuse by 45%. Containerized services with Docker and deployed on Kubernetes and GCP, reducing environment drift by 50%; built Jenkins and GitLab CI pipelines that cut release cycle time by 35%. Implemented OAuth, RBAC, centralized secrets management, and audit logging to support financial compliance requirements and pass recurring SOC 2 audits without findings.