Sep 2022 – Apr 2026 · 3 yrs 7 mos

• Supported the org developing Meta’s AI networking infrastructure, with a focus on building and scaling high-performance networks for large-scale AI clusters. Key responsibilities include delivering network capacity, operationalizing infrastructure for reliability, and ensuring an optimal performance. This work spans innovation in communication libraries (e.g., NCCL), designing and operating both scale-up and scale-out network architectures (NVIDIa GB200, GB300, AMD etc) and driving interoperability with cloud vendor compute.

Sep 2020 – Oct 2022 · 2 yrs 1 mo

• Building scalable backend to enable easy deployment and operation of complex networks that handle massive amount of data.

Oct 2019 – Jul 2020 · 9 mos · Bengaluru Area, India

• Worked on building a solution improving first-mile last-mile connectivity solution to and from mass commute hubs. The algorithm and the product focuses on optimising the cost, convenience and reliability of the first and last mile commute.

Oct 2015 – Sep 2019 · 3 yrs 11 mos · Bangalore

• Worked on building an application aware network performance monitoring solution that can help in identifying the effects of network on an application's performance. This was a 0 -> 1 product we built to solve the unique problem of providing clear visibility on networks impact on an applications performance.

Mar 2012 – Sep 2015 · 3 yrs 6 mos · Bangalore

• The Cisco Wide Area Application Services supports HTTP traffic optimisation and I worked on optimising HTTP traffic by designing and implementing features including caching and prefetching.

Mar 2008 – Dec 2010 · 2 yrs 9 mos · Chennai Area, India

• My work was majorly around improving the key performances like channel change performance, network stack performance and boot up performance of the set top box.

Dec 2010 – Dec 2011 · 1 yr · Chennai Area, India

• Developed a high-performance HTTP caching proxy that leverages multicore architecture to the highest level of parallelisation to deliver low latency & high throughput data delivery.

Jun 2005 – Mar 2008 · 2 yrs 9 mos · Thiruvananthapuram Area, India

• Implemented various healthcare applications by leveraging the parallel computing architectures using technologies like Intel SSE instructions and GPU based parallel computing.