Keval Thakarar

About Keval Thakarar

Passionate about building high-performance, scalable backend systems, I specialize in Node.js, C/C++, and TypeScript with expertise in asynchronous programming, event-driven architectures, and database optimization and distributed systems. AI Colleague : Building Agentic Operating Protocol (AOP), enabling business users to define workflows in natural language. Learn more: https://leena.ai/Agentic-AI-Architecture Libekio – Designed and developed an asynchronous event loop library supporting io_uring, epoll, and kqueue (BSD) for ultra-efficient non-blocking I/O and timer events in C. Audit Logs & Observability Platform: Architected a high-throughput distributed API interception layer and enabling real-time observability reducing L3 incident tickets. Job Recommendation System: Built an end-to-end job recommendation system leveraging semantic search with vector similarity and RAG style retrieval to match candidates against 50K+ positions. JSON Stream Parser – Built a high-performance streaming JSON parser for handling large-scale datasets (100GB+), reducing memory overhead and improving real-time data processing. Recruitment Service – Built a full-cycle recruitment platform handling user onboarding, screening workflows, and assessments with 99.9% uptime.interested Architecting distributed systems with efficient event-driven I/O and high-throughput data processing.

Experience

1. Software Engineer

   Leena AI

   Jun 2025 — Present · Gurugram, IN

Education

• Dhirubhai Ambani University

  ICT, information and communication and Technology

  2018 — 2022