vishesh chaitanya

I design and build high-throughput backend systems in Java and Spring Boot, with a focus on event-driven architectures, real-time data pipelines, and distributed system performance.

At QuilrAI, I own the core data ingestion and analytics infrastructure for a multi-tenant security compliance platform, including a Kafka-based event pipeline processing behavioral signals at scale, a Redis caching layer driving 80% cache hit rates and sub-second P95 latency, and a ClickHouse analytics engine with materialized views for real-time risk score computation. I led the migration from Neo4j to PostgreSQL + ClickHouse, eliminating ~$36K in annual licensing costs while improving query performance by 120x.

I think in terms of tradeoffs: partition strategies, consumer group semantics, schema evolution, KEDA-based autoscaling, and operational observability.

I've also built and shipped spec-agent, a Python CLI on PyPI that auto-generates Obsidian knowledge wikis from git commit history, with pluggable LLM backends including Claude, Ollama, and Groq, part of a broader interest in developer productivity tooling powered by LLMs.

Currently interested in distributed systems, platform engineering, and backend infrastructure roles at product-focused companies.

Stack: Java 21 · Spring Boot 3 · Apache Kafka · Redis · ClickHouse · PostgreSQL · Docker · GCP · Python