May 2022 – Sep 2024 · 2 yrs 4 mos

• Leading infra efficiency team at Meta, supporting multiple horizontal efficiency efforts, domain expert in deployment efficiency efforts around containerization, hardware replacement efficiency, load balancing, storage buffer management, compression
• Previously - Manage and technically lead team focusing on Performance engineering in Capacity Engineering and Analysis (CEA) org for major products at Meta, including Instagram, Messenger, WhatsApp, Video. Also, aspects of cloud demand management and efficiency.

Nov 2019 – May 2022 · 2 yrs 6 mos

• Infrastructure engineering at Meta
• Current focus - Efficiency tracking and improvements @Meta - includes fleet wide efficiency aspects:
• monitoring and reporting different aspects of utilization, software efficiency and regressions, hardware efficiency, including how to make them actionable
• Tech led and devised systematic detection of software regressions from load test of production applications, improve quality and accuracy, landed multimillion $ savings
• initiated systems for visibility of fleet-wide efficiency opportunities with quantifiable impact
• Motivating efficiency and performance through company-wide performance education, created trainings
• technically lead xfn teams to deliver efficiency tools, data pipelines
• communicate from engineers to Directors/VPs on overall efficiency state.
• Previously, Drove Capacity planning and Efficiency initiatives of Meta infrastructure for Messenger. Landed deployment re-architecture of how users are mapped to regions to optimize for efficiency and capacity fungibility.
• Improved HHVM performance through ML based approach and autotuning of performance parameters. Paper under submission. Multi million $$ savings.
• Capacity planning and performance improvements during increased load during COVID 19.

Dec 2017 – Nov 2019 · 1 yr 11 mos

• Led and managed projects in the High performance computing and 3D visualization innovation at STIC center. In addition to my technical hands-on work on HPC projects on GCP cloud, I was also responsible for:
• technically lead and prioritize innovation projects
• work with business stakeholders to ensure alignment
• work with silicon valley ecosystems - startups, Google, NVidia, universities (Stanford, Berkeley) to ensure we can benefit from cutting-edge advancements in computer science
• present HPC vision and innovation projects to customers, stakeholder, partners
• grow the team, hence recruiting and interviewing
• One of my major accomplishments has been the successful handover of cloud remote visualization prototype done at STIC to a tech center (https://www.slb.com/campaigns/gaia.aspx). The project has been commercially being offered to clients as part of our DELFI visualization experience. I also presented it at GCP booth at Supercomputing conference 2018 and Nvidia GTC conference 2019. The success outcome is handover of POCs to business stakeholder. Key successes include cloud-native 3D viz, Other key successful initiatives include HPC in Cloud with GKE, 100X faster production optimization, next gen oilfield visualization with game engines.
• I have grown the team from 2 to 6 full time people, for summer of 2019, I recruited 4 interns as well. The team has delivered:
• highly parallel and scalable production network optimizations using the Google cloud Kubernetes engine
• applying game engine technology to oilfield visualization, technologies such as Unity and MapLarge
• prototyped next generation visualization technologies such as NVidia Index, webRTC, point clouds for seismic visualization
• HPC technologies for machine learning using frameworks such as Kubeflow, distributed Tensorflow, Impala
• technology monitoring of HPC architectures and ecosystem such as TPU, quantum computing etc.

Jan 2017 – Nov 2019 · 2 yrs 10 mos

• Applying cutting edge cloud, security, and computer science technologies to HPC application and systems at Schlumberger Software and Technology Innovation Center (STIC) in Silicon Valley. Part of Schlumberger's journey of move to digitalization, specifically moving workloads and services to cloud.
• Most recently, I have been looking at frameworks and HPC infrastructure to scale ML training and RL. I am investigating Horovod, Kubeflow, IMPALA among others.
• Architect and build REST and gRPC high performance services for seismic processing and I/O using Google cloud (GCP) from ground-up. Micro-services architecture. Dealt with:
• Performance and scalability: XX performance improvements through several optimizations - caching and parallel processing, (auto) scalability.
• Security: authentication, authorization, encryption etc
• Resilience through replication, retries, failure tolerance.
• Continuous integration and continuous deployment (CI/CD) using Jenkins and VSTS.
• Code quality through static analysis tools, cppcheck, PMD, sonar, and unit test, integration tests
• Also, drive security initiatives at the center to address security issues at infrastructure, platform, and application level in cloud - firewalls, auth, VPNs etc. In addition the following:
• architecture and implementation of the first data source for the rendering architecture, which was a proof point for the visualization architecture and allowed us to differentiate, without doubt, the cloud visualization product.
• took the initiative for a large part of the security efforts done for STIC (automated firewall scans, user scans, etc.).
• mentored interns and new hires, also did GCP admin training in Houston for cloud security
• was one of the key contributors for the collaboration with other tech centers
• made hands-on contribution to security at STIC by writing scripts for periodic monitoring and alerting of firewall related security incidents

Jan 2017 – Jan 2021 · 4 yrs · San Francisco Bay Area

• Adjunct Professor in Computer Engineering at SCU from Jan 2017. Teaching graduate level course on Cloud Computing in Winter quarter 2017: includes lectures, assignments, exams, projects. So far I have taught this class 4 times, I generally teach it in winter quarter (Jan to March) every year and update the course according to latest development in the field
• For cloud computing class:
• Got AWS and Google cloud grants for student projects
• Class of 35 graduate students, 8 projects with IEEE style project reports - helped them develop the projects and plan potential conference publications
• Designed course syllabus and material, got access to presentation slides from publishers and adapted them to meet academia and industry trends in cloud computing
• developed homeworks and mid-terms to give students both theoretical as well as hands-on knowledge of cloud computing including cloud providers like AWS, GCE and cloud technologies such as kvm, mapreduce, docker containers.

May 2016 – May 2018 · 2 yrs · San Francisco Bay Area

• Technical consulting for short-lived projects and technical advise , some of the clients I have helped:
• Sociallist - helped infuse devops process into Google App Engine app - created staging and prod versions, helped fix multiple app engine specific issues including security and monitoring (https://sociallist.io/)
• Advised as a SLURM expert to a DOAR client (https://www.doar.com/experts) - included running SLURM scripts and validating SLURM capabilities on supercomputers
• Advised JISA group on micro-service architecture and devops
• Occasionally act as subject matter expert on Deepbench and Clarity
• Technical Skillset
• Private and public cloud: Kubernetes, Docker, Mesos, SDI, KVM, Google cloud, GCS, GKE, Endpoints, Amazon EC2, Azure, Openstack
• Fullstack/Backend/Distributed: scalable microservices, Python Django, Flask, auth, gRPC, protobuf, REST, swagger
• Security: Host security - SELinux, iptables, Host IDS, Suricata, Web firewall WAF, Intel TxT, Network security - IPSec, OpenVPN, Regex, IDS. Application security - static and dynamic analysis. Container security
• Perf optimization, HPC, MPI, OpenMP, shell scripting, C/C++, VTunes
• HPC for ML using Kubeflow, Tensorflow, Impala
• Misc: Zabbix, ElasticSearch, Logstash, Kibana (ELK), PubSub, AMQP, CI/CD, Git, Scrum, SLURM.

Jul 2014 – Jan 2017 · 2 yrs 6 mos · San Francisco Bay Area

• One of my key accomplishments was Intel CIT 3.2 Docker container integrity. I was the technical lead on it, I architected and prototyped the solution and later led a team to deliver the product. The software has since then open sourced (https://github.com/opencit/opencit/wiki/Open-CIT-3.2-Product-Guide#60-image-integrity-vm-and-docker)
• My other accomplishments were
• Research and develop solutions for cloud and data center security - data security and network security including SDN/NFV and SDI
• Architect and perform pathfinding for trusted computing in cloud - such as Docker containers and VMs in openstack clouds
• Trusted Docker containers, integration with Openstack
• Feature awareness in Mesos, Kubernetes, Docker Swarm
• H/W assisted performance acceleration of critical security workloads such as VPN, IPsec

May 2012 – Aug 2012 · 3 mos

• Researched techniques for application-aware VM placement in cloud (mentor Dr.
• Dejan Milojicic)

Sep 2011 – Sep 2013 · 2 yrs

• Designed and implemented methods for bridging HPC-cloud divide. Presently, collaborating on performance evaluation and simulation of next-generation systems for data-intensive applications

May 2011 – Aug 2011 · 3 mos

• Evaluated the performance and mapping of HPC applications in cloud (mentor
• Dr. Dejan Milojicic)

Aug 2009 – Jun 2014 · 4 yrs 10 mos · Urbana-Champaign Area

• Worked on various projects: large-scale HPC applications, parallel runtime systems, and schedulers for both clouds and HPC (select projects lised below).

Jul 2008 – May 2009 · 10 mos · Greater Hyderabad Area

• Worked in the team Microsoft CRM (Customer Relationship Management)

May 2007 – Jul 2007 · 2 mos · Bangalore

• Worked on Apache Tomcat server deployment
• Worked on multi-threaded SOAP request generator