Backend Software Engineer
I build scalable Python backends — high-throughput distributed pipelines, event-driven microservices, and cloud-native APIs. Currently shipping production systems at WebMD that process millions of documents with multi-provider AI integrations.
From greenfield APIs to high-throughput distributed systems — I focus on correctness, performance, and systems that hold up in production.
FastAPI microservices and RabbitMQ-backed pipelines designed for production load. Batch processing, bulk DB operations, and efficient write strategies that scale without rewriting.
PostgreSQL schema design, query optimisation, and vector storage (pgvector). Redis caching and pub/sub. Multi-backend routing so you're not locked into one store.
Multi-provider LLM routing (OpenAI, Azure, Anthropic, Vertex, Cohere) and embedding pipelines wired into production backends. I integrate AI as a service, not a research project.
PII/PHI detection and anonymisation using Presidio + spaCy. HIPAA-aware service design. OAuth2, JWT authentication, and secure API patterns from day one.
RabbitMQ consumers, idempotent processing, deduplication strategies, retry + DLQ patterns, and queue pressure controls — systems that handle failures gracefully.
Prometheus + Grafana dashboards wired to latency, throughput, and error metrics. Docker packaging, Kubernetes deployments, and CI/CD pipelines that ship confidently.
Production engineering across healthcare technology and fintech.
Production-grade personal projects demonstrating distributed systems design and AI integration.
Production-grade multi-channel notification platform built for extreme throughput. Per-channel queue isolation, exponential backoff retry, Dead Letter Queue, and full observability.
Agentic RAG system for investigative research with multi-step tool-using agents, FAISS vector search, and automated hallucination detection.
Tools I use in production, not just on a résumé.
Open to full-time remote roles, contract work, and interesting backend problems worldwide.