Abhishek Pandey

Software Engineer

San Francisco, California, United States13 yrs 3 mos experience
Highly StableAI Enabled

Key Highlights

  • Led cross-team collaborations at Meta for AI/ML infrastructure.
  • Successfully migrated 7B+ records at Netflix, enhancing system performance.
  • Developed fault-tolerant systems at Uber, improving reliability.
Stackforce AI infers this person is a Backend-heavy Infrastructure Engineer with expertise in high-scale distributed systems.

Contact

Skills

Core Skills

Machine LearningDistributed SystemsSoftware Engineering

Other Skills

AI/ML Infrastructure DevelopmentPyTorchTritonHardware AgnosticPreprocessing StackRPC solutionsGRPCGraphQLRESTOpen SourceWeb ClientMicroservicesAPIsData ModelingCollaboration

About

Building AIML foundational infrastructure for billions of Meta users around the globe. I have led teams and done cross-team collaborations across large organizations.

Experience

13 yrs 3 mos
Total Experience
2 yrs 7 mos
Average Tenure
2 yrs 8 mos
Current Experience

Meta

Software Engineer

Sep 2023Present · 2 yrs 8 mos · Menlo Park, California, United States · Hybrid

  • AI/ML Infrastructure Development (PyTorch PreProc). Making preproc hardware agnostic by standardizing on PyTorch Native and Triton kernels, the new architecture delivers a developer-friendly, high-performance, and hardware agnostic preprocessing stack for both training and inferencing stages of ML models.
AI/ML Infrastructure DevelopmentPyTorchTritonHardware AgnosticPreprocessing StackMachine Learning+1

Netflix

2 roles

Senior Software Engineer, Developer Platform - RPC Infrastructure

Feb 2022Sep 2023 · 1 yr 7 mos

  • The team enables Netflix developers with RPC solutions using GRPC, Domain graph service (GraphQL), and Web client (REST) based communication. We also have huge presence in open source community and our solutions are utilized by developers around the globe.
  • Enabled web client for external commerce protocols to get Ads data securely and reliably. This was critical to launch Netflix’s Ads tier globally.
  • Migrated and Integrated IPC clients to service mesh.
RPC solutionsGRPCGraphQLRESTOpen SourceDistributed Systems+1

Senior Software Engineer, User Systems and Data

Jun 2018Feb 2022 · 3 yrs 8 mos

  • The team builds distributed systems and infrastructure around all the user’s data, state, identity, preferences, experience level, and device relationships on a web-scale.
  • Designing and implementing ​high-performance​ large-scale distributed ​tier-0 microservices, ​APIs, data modeling​, and​ systems infrastructure​ for the ever-growing Netflix's subscribers need as well as tuning our products for web-scale.
  • Hiring, mentoring, and building a stellar engineering team.
  • Notable projects:
  • Introduced the new account type in the Netflix ecosystem for special purposes.
  • Lead, architected, and implemented the new account type, the orchestration layer, and worked with almost every engineering team, internal and external partners, vendors, security, privacy for this huge multi-org effort.
  • Rewrote existing subscriber/user systems and migrated 7B+ records from the old system with maintaining active reads and writes and ensuring the records are intact and consistent. Huge impact on the latency, throughput, ease of releasing features, and saving millions of dollars for Netflix. Worked with 100+ partners and services for the seamless migration effort.
Distributed SystemsMicroservicesAPIsData ModelingSoftware Engineering

Uber

Software Engineer

Jun 2016Jun 2018 · 2 yrs · San Francisco, California

  • Uber's Dynamic Configuration Infrastructure team, that store configs and serve them dynamically across all the hosts in Uber Datacenter. You get those configs from memory in your service in order of nanoseconds while evaluating elaborate rules around the context of the configs.
  • Developed highly distributed systems, fault tolerant tier-0 microservices at scale that run Uber dynamically.
  • Drove cross org projects to help run Uber as reliable as running water.
  • Lead the new grad interview panel at Uber. Interviewed 100+ new grads from across the universities and helped them learn more about Uber engineering.
  • Notable projects:
  • Configuration automated rollback in case of Uber outage caused by a bad change. Led the uber wide initiative comprising of multiple teams from monitoring uber health to triggering crucial actions. High impact in elevating Uber's availability.
  • Dynamic configuration caching/gateway layer to serve configuration across all the hosts in the uber datacenter reliably. The caching layer served data across 50K hosts in nano-seconds and served as the fallback for the configuration backend.
  • Migrated and integrated geofences with configurations which allowed configurations to target multiple geofences globally.
  • Wrote configuration libraries, backend and made configuration GDPR, project fidelity compliant.
Dynamic ConfigurationFault ToleranceMicroservicesDistributed SystemsSoftware Engineering

Appdynamics

Software Engineer

Dec 2013Jun 2016 · 2 yrs 6 mos · San Francisco Bay Area

  • Developed products, tools, APIs and extensions for Appdynamics APM software.
  • Lead the design and development of customer monitoring application - Proactive
  • Helped customers with debugging and solving complex issues in their production environments.
APM SoftwareAPI DevelopmentCustomer MonitoringSoftware Engineering

Walmart global tech

Software Engineer

Feb 2013Dec 2013 · 10 mos · San Francisco Bay Area

  • Developed data intensive restful web-services and frameworks for next generation e-commerce platform for walmart.com.
  • Worked on item pricing and store finder web services that are used in walmart.com
RESTful Web ServicesE-commerce FrameworksSoftware Engineering

Qiagen

Software Engineer

Sep 2012Dec 2012 · 3 mos · San Francisco Bay Area

  • Developed server side java web utilities in a multi-threaded environment.
  • Added new features in the current product to support multiple access of data from different client requests.
JavaMulti-threadingSoftware Engineering

Siemens plm software

Software Engineer Intern

Feb 2012Aug 2012 · 6 mos · Milford,OH

  • Developed tools and utilities using Server - Client architecture in J2EE.
  • Developed web based internal tools using Java Servlets & JSP.
  • Wrote Perl Scripts for initializing VMs and loading start-up classes and utilities.
  • Developed XML based dashboard for upgrade
  • Developed Ant Scripts to build the application.
  • Used Unix shell scripts for backend processing.
  • Involved in development and deployment of application on UNIX environment.
J2EEJava ServletsPerl ScriptsSoftware Engineering

Education

University of Cincinnati

Master of Science (Thesis) — Computer Science

Jan 2010Jan 2012

Rajiv Gandhi Prodyogiki Vishwavidyalaya

Bachelors of Engineering — Information Technology

Jan 2006Jan 2010

Stackforce found 100+ more professionals with Machine Learning & Distributed Systems

Explore similar profiles based on matching skills and experience