Backend Engineer working on high-concurrency distributed systems and AI infrastructure

Built systems handling ~850+ req/sec with sub-500ms latency under real-world load

Open to remote opportunities and global collaboration

• High-concurrency backend systems (500–850+ req/sec under load)
• Distributed AI systems (RAG, LLM orchestration, vector search)
• Event-driven architectures (Kafka, async pipelines)
• API-first platforms for integrations and automation
• Low-latency systems optimized using caching, batching, and routing

Production-grade distributed backend system designed to handle real-world AI workloads at scale.

• ~850 req/sec throughput (load-tested)
• 500+ concurrent requests with stable latency
• 100K+ documents processed

• Async FastAPI services (stateless, horizontally scalable)
• Redis distributed caching
• FAISS vector index for semantic retrieval
• Kafka-based event pipelines
• Multi-LLM routing layer with fallback handling

• Stateless architecture for horizontal scaling
• Async pipelines for non-blocking execution
• Cache-first design to reduce latency and cost
• Backpressure handling for stability
• API-first approach for extensibility

• ~40% latency reduction
• ~30% cost reduction
• Stable under sustained and burst traffic

graph TD
    Client --> CDN
    CDN --> LB[Load Balancer]
    LB --> API[FastAPI Gateway]
    API --> Cache[Redis Cache]
    API --> Workers[Async Workers]
    Workers --> VectorDB[FAISS Index]
    VectorDB --> LLM[LLM Providers]
    LLM --> Response

• Distributed systems and system design
• High-throughput backend engineering
• AI infrastructure and LLM systems
• Performance optimization and reliability

• Backend Engineer (Distributed Systems)
• AI Infrastructure
• High-Concurrency Systems