Akanksha Mahajan

Senior Software Engineer

San Francisco, California, United States10 yrs 5 mos experience
Highly Stable

Key Highlights

  • Expert in optimizing large-scale distributed systems.
  • Significant contributions to RocksDB open-source project.
  • Strong foundation in Computer Science with a Master's degree.
Stackforce AI infers this person is a Backend-heavy Fullstack Engineer with expertise in Distributed Systems and Data Management.

Contact

Skills

Core Skills

C++Distributed Systems

Other Skills

AlgorithmsAndroid DevelopmentApache KafkaApache ZooKeeperCCassandraCilkPlusCommunicationData StructuresDebuggingDistAlgoFile SystemsGitInfrastructureJava

About

I’m an experienced SWE currently working at LinkedIn. During my time at Meta, I specialized in the RocksDB (an open source project https://github.com/facebook/rocksdb/wiki )—a dominant key-value storage engine in Meta's infrastructure and widely adopted across industries for persistent data storage and online query serving. Equipped with a Master's degree focusing on distributed systems and operating systems, and my internship at Nutanix in distributed systems, I am dedicated to leveraging cutting-edge technologies to optimize performance and scalability in large-scale distributed systems. I am passionate about driving innovation and solving complex challenges in the realm of storage and data management. Graduated with Masters in Computer Science from Stony Brook Univeristy Courses taken : Operating system, System Security and Data Mining, Distributed Systems, Analysis of Algorithms and Fundamentals of Data Science

Experience

10 yrs 5 mos
Total Experience
1 yr 10 mos
Average Tenure
1 yr 2 mos
Current Experience

Apple

Senior Software Engineer

Apr 2025Present · 1 yr 2 mos · Cupertino, California, United States

  • FoundationDB

Linkedin

Senior Software Engineer

Jul 2024Apr 2025 · 9 mos · Sunnyvale, California, United States

  • Part of Storage team

Meta

Software Engineer

Feb 2020Jun 2024 · 4 yrs 4 mos · Menlo Park, California, United States

  • RocksDB team - Open source | C++ | Key-Value Storage Engine
  • Drove optimization initiatives for the Scan module, delivering a notable 32% reduction in fast path bandwidth in the production environment. Presented my work at the RocksDB open-source meetup.
  • Implemented async I/O from scratch, including support in Iterators during Seek and prefetching, recognized and pub- lished in an ACM paper (https://dl.acm.org/doi/pdf/10.1145/3589772). Resulted in a substantial 35% decrease in scan latency and 40% in p99 metrics.
  • Led the initiative to integrate FileSystem-provided buffer utilization for all reads, eliminating memcopy overhead and yielding a 3-5% CPU improvement for Distributed Network File System (NFS).
  • Led the design and implementation of a comprehensive RocksDB release validation internally from scratch, ensuring rigorous correctness testing and performance regression analysis for new releases.
  • Developed strace-like tool for RocksDB IOs IOTracer, enhancing visibility and latency analysis.
RocksDBC++Distributed Systems

Nutanix

MTS intern

May 2019Aug 2019 · 3 mos · San Jose, California

  • Part of Core Data Path team
  • Worked on distributed data-log abstraction on top of Nutanix VDisks to extend data log on multiple VDisks by using Nutanix object storage. This project will be a groundwork for various applications on top : pub sub, stream ingestion etc.
  • Technologies involved : Zookeeper, Cassandra, RocksDB
  • Language: C++
  • Nutanix U-Hackathon : Implemeted a FUSE file system backed by zookeeper to make zeus library language agnostic by allowing services to use file system functions to access Zookeeper information on cluster.

Mentor graphics

Senior Member Technical Staff

Dec 2015Jun 2018 · 2 yrs 6 mos · Noida Area, India

  • I was part of Runtime team which develops OS for Veloce Emulator.
  • Developed new features for power estimation utilities - Velsaif and and Activity Plot. ( C++)
  • Features :
  • Activity Plot: Enabled export of raw toggle count data and introduced Max Plots to visualize the
  • highest peak value against time in a trace zone.
  • Partitioned SAIF le generation by time zone and facilitated time-windowed waveform extraction
  • and SAIF generation.
  • Re-engineered waveform reconstruction for 2-state-nets and signals to use 128-bit computations,
  • which yielded a 10% performance boost.
  • Optimizations :
  • Introduced parallelism and a save-and-restore feature for grid-launched processes using a
  • Leader/Follower model. This speed up runtime power analysis up to 30%.
  • Automated restart of power analysis processes on failure during hardware emulation, eliminating
  • any need for customer intervention.
  • Implemented compression of system-wide temporary les using zlib to reduce disk usage to 30%.
  • Re-factored cascaded if-else statements to virtual classes and templates of C++, reducing computational
  • overhead.

Samsung

Software Developer

Jun 2014Nov 2015 · 1 yr 5 mos · Noida Area, India

  • Part of System Application team (Call settings, Dialer and Contacts app) for UI/GUI developement, new feature implementation and OS Upgrade for all the latest Android Samsung flagship devices and tablets.

Education

NIT Jalandhar

Btech — Computer Science

Stony Brook University

Master's degree — Computer Science

Stackforce found 100+ more professionals with C++ & Distributed Systems

Explore similar profiles based on matching skills and experience