Karthick Pattabiraman

Designed and developed core modules of the Schedule Management service — GraphQL APIs, shift management, and the V2 Booking schema now powering scheduling across 2 major product lines via a company-wide API federation.

Eliminated per-appointment authorization overhead in the Calendar Segments API by replacing N sequential auth calls with a single batched call, cutting response time by ~70%.

Built a real-time Apache Flink pipeline fanning out appointment events to 4 downstream systems with standardized retry and error-handling contracts across all Kafka producers.

Led the Java 21 and Elasticsearch 8 migration across 6 services, replacing deprecated clients and resolving security vulnerabilities.

Architected MCP servers for the Appointments and Expert Matching domains, integrating with AI models to enable real-time scheduling queries and expert matching — adopted by 3 internal teams within the first month.

Executed chaos engineering experiments across 6 services — pod deletion, network fault injection, CPU/memory stress testing, and DNS fault tolerance — maintaining compliance with 30-day rolling resilience certification.

Built 3 microservices for a third-party vehicle access gateway — Spring OAuth2, VIN-to-vehicle-ID mapping, and JWT token exchange securing ~800 monthly in-car delivery sessions for partners.

Architected the Kafka-based command pipeline routing voice-to-vehicle commands (lock, unlock, remote start) from smart assistants, processing 50,000+ commands daily with sub-second p99 latency.

Implemented Redis caching for vehicle state lookups (door status, location, fuel level), cutting upstream Connected Vehicle Cloud calls by ~60% and reducing response times from 1.2s to under 400ms.

Owned the real-time telemetry ingestion pipeline — Spring Boot consumers processing ~2 million vehicle sensor events/day from Kafka to BigQuery, with Python ETL scripts handling data cleansing and anomaly detection.

Added Resilience4j circuit breakers after running FMEA across 5 failure scenarios, reducing cascading failures by ~40% and maintaining 99.9% API availability with p95 latency < 800ms.

Partnered with a major social platform's developer team on a beta integration program to add social media as a booking channel — drove a 45% increase in Booking Page traffic and contributed ~$90K in ad-driven revenue.

Built third-party platform integrations — video conferencing for appointments and a native booking flow inside a call-center platform that improved agent productivity by 50% and cut call costs by 20%.

Eliminated Booking Page API bottlenecks through Redis caching, query optimization, and datastore index tuning, reducing response times by 65%.

Migrated the data layer to support micro-frontend architecture, transforming and validating ~3 million records with Python — enabled 2 new brand launches on the same codebase and sped up release cycles by 70%.

Led a backend API team of 4 — code reviews, mentorship, and release ownership across the scheduling product line.

Reduced signup API latency by 80%, directly improving new customer onboarding conversion.

Led the migration from GCM to Firebase Cloud Messaging, improving push notification delivery reliability to 98%.