Published on
Posted 10 days ago

ML Engineer [IC3]

Authors

Job Details

Role: ML Engineer [IC3]

Location: San Francisco (Remote)
Job Type: Full-time

About the Role

Company Overview:
Sourcegraph’s mission is to empower everyone to code by building the most powerful code intelligence platform in the world. Sourcegraph has raised 225Mwitha225M with a 2.625B valuation from prominent investors like a16z, Sequoia, and Redpoint. Their goal is to automate 99% of the toil in software engineering using their code AI tools, leaving the remaining 1% to human creativity.

Sourcegraph’s customers include top companies such as Uber, Plaid, and 4/5 FAANG companies, among others. The company operates with an all-remote workforce but prefers candidates based in San Francisco for this role, with at least 20 hours of overlapping work time in Pacific Standard Time (PST).

As a ML Engineer [IC3], you’ll contribute to building out Sourcegraph’s machine learning capabilities, particularly in the context of AI-powered coding assistance. You will work on end-to-end ML pipelines, from development to production, scaling infrastructure, and driving innovation in LLMs.

Why This Job is Exciting

Sourcegraph’s code intelligence platform offers unparalleled context for large language models (LLMs). By joining their ML team, you will help advance the world’s most powerful coding assistant and build cutting-edge tools that multiply developer productivity.

This role provides opportunities to push the boundaries of AI and ML in the developer ecosystem. Additionally, Sourcegraph’s enterprise distribution pipeline allows engineers to deploy their innovations directly to some of the largest code bases in the world.

Your Responsibilities

  • Build feature computation, storage, monitoring, and serving systems for Sourcegraph’s Cody LLM stack.
  • Develop distributed training infrastructure and backend services for high-QPS, low-latency use cases.
  • Prototype and experiment with Cody prompts and context sources to improve AI models.
  • Own key pieces of Sourcegraph’s ML infrastructure and drive technical vision for major ML components.
  • Design and implement a high-QPS inference service using CPU/GPU resources.
  • Conduct research and stay current with relevant advancements in ML and AI.

Required Qualifications

  • Experience:
    • 5-8 years of experience in ML engineering with a focus on end-to-end ML pipelines.
    • Proven expertise in deploying ML models to production and working on the full ML lifecycle.
    • Demonstrated ability to optimize ML models and experiment with offline/online metrics.
  • ML Tools and Infrastructure:
    • Experience with distributed backend services and scaling ML infrastructure for production use cases.
    • Ability to develop feature pipelines and deliver high-QPS, low-latency systems.
  • Collaboration:
    • Working hours overlapping with PST (Pacific Time) for at least 20 hours per week to collaborate with the team.

Compensation and Benefits

Sourcegraph offers above-market compensation, aiming to attract the best talent worldwide.
Salary: $185,000 USD base
Sourcegraph’s compensation structure takes into account the cost of labor index for different regions, and their transparent pay bands are available to all employees.

In addition to cash compensation, Sourcegraph provides equity options, generous perks, and benefits designed to keep employees happy and healthy.

Interview Process

The interview process at Sourcegraph is thorough and involves several stages, designed to assess technical depth, collaboration, and cultural fit:

  • Introduction Stage:
    • [30m] Recruiter Screen
    • [45m] Technical Deep Dive
  • Team Interview Stage:
    • [60m] ML Depth Interview
    • [60m] ML Breadth & ML Systems
    • [15m + async] Pairing Exercise
  • Final Interview Stage:
    • [30m] Values Interview
    • [30m] Leadership Interview with co-founder

You can learn more about Sourcegraph’s interviewing philosophy in their handbook.

About Sourcegraph

Sourcegraph is committed to building an inclusive and diverse workplace and welcomes applicants from all backgrounds. The company participates in E-Verify for U.S. employees and offers competitive compensation along with practical benefits to keep the team at their best.

Learn more about Sourcegraph by visiting their handbook.