Open to opportunities

Rajat Raj

@rajatraj1

Message

Backend Engineer (SDE II) building AI-powered distributed systems and LLM integrations.

India

Message

What I'm looking for

I want to build cloud-native backend systems that blend real-time data pipelines with LLM integration. I’m looking for teams that care about low-latency performance, correctness (strong semantics), and production reliability with strong observability.

I’m a Backend Engineer (SDE II) and Competitive Programmer who builds AI-powered platforms and scalable distributed systems. I focus on turning complex data and real-time requirements into reliable, low-latency services—end to end.

At SuperAGI, I’ve led schema-driven platform work and microservices migration, decomposing a legacy Rails monolith into independent Go services. I’ve delivered exactly-once bulk processing with Kafka, built fast filtering/search with OpenSearch, and created an AI-powered alerts service using OpenAI/Anthropic with fuzzy deduplication.

Previously at Oracle Cloud Infrastructure, I designed and operated backend systems for Java runtime management across ~1M deployed agents. I improved security by automating URL rotation, built monitoring/alerting that reduced MTTD to under 5 minutes, and owned incident response including RCA for multiple Sev1 events.

I’m energized by practical engineering—LLM orchestration, observability, and production-grade reliability. I also bring a competitive-programming mindset (CodeChef top 1% in 2025) that helps me think in constraints, performance, and correctness.

Experience

Work history, roles, and key accomplishments

Current

Software Development Engineer II

Current

SuperAGI

Sep 2025 - Present (9 months)

Designed and built a schema-driven CRM platform in Go, enabling dynamic objects and high-throughput operations while migrating 10M+ records with zero downtime. Reduced memory 40% and p99 latency 2x by decomposing a Rails monolith, and delivered Kafka/Flink exactly-once bulk processing plus real-time CDC into ClickHouse and OpenSearch search at ~300ms latency.

Go Kafka Flink Clickhouse Redis OpenSearch Kubernetes OpenTelemetry OpenAI API

Member of Technical Staff

Oracle Cloud Infrastructure

Jul 2023 - Sep 2025 (2 years 2 months)

Built a gRPC-based Java agent management system covering ~1M deployed agents and expanded adoption by 50% by supporting Oracle JDK 8 Enhanced Performance Packs and OpenJDK distributions. Improved security and reliability by rotating pre-authenticated storage URLs (14 days to <1 hour), automating global ETL across 95+ regions (5 days to 6 hours), and reducing mean time to detection from 30 minutes