Itisha Dewan Email: ishadewan07@gmail.

com
Linkedin Profile Mobile: +1-857-425-9389
Experience
Google Sunnyvale, CA, USA
Software Engineer II Feb 2023 - Present
• Contributed to the design and development of data integrity pipelines around deletion systems in C++ for a new
user-data storage infrastructure being built to reduce the total byte storage overhead for Blobstore and
Google Cloud Storage (GCS).
◦ Collaborated with multiple teams on adding backward compatibility support for existing jobs to ensure
durability and restorability requirements are met in the new storage system.
◦ Extended recovery jobs to support the new system and enforce compliance with data retention and storage
policies, accompanied by extensive unit and e2e testing.
◦ Authored documentation for API changes and developer instructions into a comprehensive guide covering the
integration of the new storage system with the data integrity framework.
• Worked with a team on the design and implementation of a new Garbage Collection system for Blobstore and GCS,
which aims to significantly improve the efficiency and scalability of some stages in the garbage collection
process.
Meta Menlo Park, CA, USA
Software Engineer II Feb 2019 - Jan 2023
• Successfully launched a large-scale infrastructure migration for the Workplace chronological feed
involving complex architecture re-design in C++.
◦ Achieved 100× latency improvements in loading the chronological Feed as well as other performance
improvements. This also unlocked opportunities for latency improvements in all other Workplace Feeds.
◦ Led a team of engineers through gathering requirements, aligning on the design, proof-of-concept, capacity
estimation, execution, experimentation, testing and safe staged roll-out.
◦ Allowed for custom configuration of features and filters for the Workplace product team and isolating the
individual feed against unintended changes causing unintended behaviour and/or regressions.
• Optimized Watch Feed tab in the Facebook App with 2% CPU time improvement (stat-sig) by improving
the backend infrastructure using C++ and Hack PHP. I drove the execution, investigations and statistical
experimentation to prove that the optimization was significant and this was then shipped to 1.98B DAU of the
Facebook App.
• Led the effort to build Custom Feed platform in C++ which was used to migrate the Facebook Most
Recent Feed to Instagram Machine Learning (IGML) backend infrastructure.
◦ Gathered and aligned on the requirements and success metrics with multiple team stakeholders.
◦ Studied various parts of the existing Feed stack and built tools to identify and fill gaps in the new platform.
◦ Studied the existing components of IGML infra and built from scratch, the tooling and logic for the migration.
◦ This improved developer efficiency for the product team, and greatly simplified and modernized the Most
Recent Feed tech stack, while maintaining behavioral and performance parity. This was also the first step in
migrating all feeds (other than the main FB app feed) to IGML infra and provided proof-of-concept that other
Feeds could be migrated to this infra.
• Built an internal tool using Python to monitor and clean our critical Warm Storage for all Feed request
data which is used for gathering product level metrics for Feed along with incident debugging and health
monitoring. This was technically complex and critical to ensure the availability of data in our storage.
Meta Menlo Park, CA, USA
Software Engineering Intern June 2018 - Aug 2018
• Worked with the Commerce Search team to predict a feature for Marketplace products using primarily regex, which
improved the conversion rate of these products within the Marketplace experience.
Openstack Remote
Outreachy Intern May 2016 - Aug 2016
• Worked on the OpenStack open source project to migrate Glance (OpenStack Image Service) Replicator to using the
requests library from the httplib library, and added a command line option to specify a custom certification bundle.
Education
Boston University Boston, MA, USA
MS in Computer Science with Specialization in Data Centric Computing; GPA: 3.61 Sept 2017 - Jan 2019
Courses: Machine Learning, Artificial Intelligence, Advance Databases, Data Mining
International Institute of Information Technology Hyderabad Hyderabad, India
B.Tech. in Computer Science and Engineering; GPA: 7.24 July 2013 - May 2017
Courses: Information Retrieval and Extraction, Database Systems, Operating Systems, Data Structures, Algorithms
Technical Skills
C++, C, Python, Hack PHP, BASH, Mercurial, Git, HTML5+CSS, Web2Py, Django, MySQL, Gerrit