Omkar Katta

I’m a backend engineer with 3 years building high-throughput Django/Python services, data platforms, and developer tooling. I’m driven by system design and measurable performance improvements, and I enjoy turning reliability and latency goals into clean, scalable implementations.

At UniAcco, I shipped APIs handling thousands of concurrent leads and reduced production latency 7.5× by rewriting ORM queries and layering Redis caching on high-traffic endpoints. I also improved developer throughput by shrinking Docker images (2.5GB→900MB) and cutting CI/CD build time (2m40s→1m30s) using GitHub Actions Docker layer caching. On the data/ops side, I built a bulk API with atomic transactions, Celery async workers, and S3 uploads, and I designed an Airflow orchestration platform with reusable DAG templates to reduce ETL deployment lead time by 40%.

Previously at Brane Enterprises, I increased data processing throughput by 20% by re-architecting Django REST APIs with optimized PostgreSQL queries and SQLAlchemy connection pooling. I reduced API error rate by 35% through a service-layer abstraction that decoupled business logic from I/O, and I sped up team execution by standardizing Git branching, PR review checklists, and automated CI/CD pipelines.

As a Data Engineer at Rapido, I improved analyst model accuracy by 15% with an end-to-end Data Health monitoring feature using Python, Spark, Airflow, and PostgreSQL. I also extended a SQL parser to support multiple CTEs for better query efficiency (30%) and deployed the platform on Kubernetes to achieve a 99.9% uptime SLA. Outside core delivery, I published an open-source “route-guard” Claude Code plugin that enforces deterministic token-budget governance to prevent runaway token consumption.