Jun 2023 – Oct 2025 · 2 yrs 4 mos · Bengaluru, Karnataka, India · On-site

• Architected a deep learning pipeline for 5G PRACH detection using a noise-augmented dataset, outperforming conventional methods by 4 dB while maintaining a false alarm rate below 0.1%
• Implemented the full inference stack in C++ with AVX-512 optimizations, achieving sub-5ms latency and enabling real-time deployment on production hardware
• Redesigned and optimized PRACH modules on the VSPA architecture, delivering a 10× improvement in timing performance while guaranteeing production-grade reliability.
• Built Python-based automation frameworks for large-scale log analysis, accelerating debugging cycles and significantly increasing developer productivity
• Collaborated with cross-functional teams to integrate PRACH into end-to-end 5G platforms, ensuring standard compliance and smooth system rollout.

Jun 2022 – Aug 2022 · 2 mos · Hyderabad, Telangana, India