Jun 2024 – Jan 2025 · 7 mos · Remote

• Developing and improving the core compute platform as part of the Linode team, adding features and increasing reliability.

Sep 2021 – May 2024 · 2 yrs 8 mos · Bengaluru, Karnataka, India

• Implemented automated vulnerability patching and application restarts with zero downtime across FK systems
• reducing patching time from 2-3 months to 1 day. Presented talk in 24th Debian Conference and India FOSS 3.0.
• Talk link: https://www.youtube.com/watch?v=pwd4gB8xvas
• Created end to end automated image testing framework and pipeline for creation of Debian base image to be used
• across the org in GCP and FK Data center. Helped reducing image release lifecycle to single click deployment.
• Built longevity testing framework detecting hardware and scalability issues on baremetal servers at large
• scale.Scraped an ansible based project due to scalability issue and built our own solution in bash.
• Created a VM lifecycle library as well as cli tool used in various projects & developers saving hours of developer
• time monthly.
• Migrated DNS service seamlessly with minimal downtime across both Data center actively serving thousands of users.
• Migrated a libvirt based VM controller service from python2 to python3. Refactored and added debian package
• support for the project.
• Migrated Jenkins deployment with seamlessly from one DC to another without user downtime.
• Improved network bandwidth utilization while Live Migration of virtual machine.

Feb 2019 – Sep 2021 · 2 yrs 7 mos · Bengaluru Area, India

• BPF, XDP(high performance network data path in-kernel). Kernel networking, tracing.
• bpftrace(https://github.com/iovisor/bpftrace), pbench(https://github.com/distributed-system-analysis/pbench/)

Aug 2018 – Jan 2019 · 5 mos · Bengaluru Area, India

• Working on Redhat Satellite perf and scale. Puppet 5 performance, alternative memory allocator analysis.
• Working on XDP(an ebpf program) performance. XDP(express data path) is a network data path in the Linux kernel that provides early packet processing even before the allocation of socket buffer enabling it to perform faster while still being programmable via ebpf(which allows runtime loading of program without rebooting the kernel).

Mar 2018 – Aug 2018 · 5 mos · Remote

• Debian project

Mar 2017 – Jun 2018 · 1 yr 3 mos · New Delhi Area, India

• Organising monthly meetups of India Linux user's group, Delhi chapter.

Jan 2017 – Apr 2017 · 3 mos

• Built back-end for the product.
• Implemented caching using Memcached
• Setting up cronjobs and wrote tests.
• Wrote script to extend a Deeplearning model for videos and used multiprocessing to speed up process-
• ing.
• Tech stack:
• Languages - Python,Javascript,HTML,CSS
• Framework - Django/Python
• Testing - Unittest
• Tools Used:
• Rabbitmq(task queueing)
• Celery(asynchronus tasks)
• Docker(running Caffe image)
• ffmpeg
• Memcached (caching)
• Database - PostgresSQL

Nov 2016 – May 2017 · 6 mos

• This initiative allows people to have access to influential research papers in various areas of computer science including crypytography,AI,comp-vision. By working in close collaboration with other scholars, short length talks 5-10 min duration is created. This initiative is carried out under the guidance of Dr.Rajan Vaish, who is a Post-Doc Researcher at Standford University.
• Worked on the papers:
• Wander Join: Online Aggregation via Random Walks. (https://www.cse.ust.hk/~yike/sigmod16.pdf)
• Compressed Linear Algebra for Large-Scale Machine Learning. (http://www.vldb.org/pvldb/vol9/p960-elgohary.pdf) - DRI (Directly responsible individual) for this talk.
• http://scholar.stanford.edu/