Backend engineer with 5 years of experience building scalable Java/Spring services on AWS, with recent focus on applied AI products and production LLM integrations.
Owned enhancements to a Spring Boot pricing service with ScriptSave integration, delivering nearby-pharmacy and item-detail APIs supporting 200k+ procedures/drugs and 10M+ regional price points while reducing member price-lookup escalations.
•
Co-led backend delivery for v1 of a generative AI member chatbot using AWS Bedrock and Python FastAPI, designing retrieval and service integration across internal knowledge and live backend APIs; achieved p95 <2.5s, 25–30% containment, and ~18% lower support AHT while meeting HIPAA expectations.
•
Built a real-time AI call-summary system integrating Genesys Cloud transcripts into an internal admin platform, eliminating manual post-call documentation and saving agents ~5 minutes per call across 200–300 daily interactions (~20+ agent-hours/day).
•
Built an n8n workflow to auto-triage prod tickets by matching similar Jira issues via JQL, analyzing related GitHub commits, and generating PR-ready fixes for on-call engineers; reduced triage time by ~40% and enabled ~30% of recurring tickets to be resolved with approval only.
•
Designed and operated a Spring Batch backend for actuarial and pricing uploads with validation and S3-backed workflows, sustaining ~1.2M rows/hour at 94.5% successful runs.
Spearheaded a Python/FastAPI backend service connected to an internal analytics engine, enabling real-time predictive insights and correlations across 100k+ industrial attributes.
•
Built 3 WebFlux streaming endpoints that reduced service memory usage by >90% and improved concurrent throughput for backend consumers.
•
Integrated Aveva PI Server via Spring Boot, importing 130k+ assets/attributes into PostgreSQL and enabling scheduled ingestion of 1M+ points to support downstream services and analytics.
•
Built high-throughput backend data flows with Flink + Kafka and Telegraf-to-InfluxDB, processing 5M+ records and 100GB+ of time-series data as infrastructure scaled.
Built a Spring Boot backend for a loan-approval system integrating Confluent Cloud Kafka, Amazon MSK, and custom Avro; automated applicant verification via external APIs and processed 11,000+ applications in 3 months.
•
Shipped 10+ REST microservices using AWS (S3, MSK, DynamoDB, CodePipeline) and LaunchDarkly feature flags to support controlled rollout and efficient delivery.
Designed and developed internal trading and operations applications using Java/J2EE, Spring Boot (MVC & Batch), React with TypeScript/Redux, MySQL, and AWS (API Gateway, Lambda, S3) in a high-performance enterprise environment.
•
Optimized Spring Batch tax-document generation, improving processing time by 18%, and expanded Java ETL support to 11 transaction types, including equities and options.
