About the work

GitLab ModelOps aims to bring data science into GitLab both within existing and new features to make them smarter, more intelligent, and to empower GitLab customers to build and integrate data science workloads within GitLab. This stage also supports our FY24 Product Theme: GitLab for Data Science, and the GitLab company vision of developing an AllOps platform.

Your role will be to build and optimise our infrastructure and scale our LLM’s (large language models) while integrating them as microservices within Gitlab, and dropping the costs of Machine Learning. You will balance short and long-term efforts from an infrastructure perspective and work on scalability challenges. You will be the backbone of our ML efforts that solve the challenges of today and tomorrow.

 Some examples:

• In GitLab version15.4 we released our Suggested Reviewers Open Beta
• In Gitlab version 15.3 we released Code Suggestion Closed Beta:  a short blog

What you will do

• Play a key role in the design, implementation, and integration of product features;
• Solve technical infrastructure problems of high scope and complexity;
• Test, deploy, maintain, and improve ML infrastructure and software that uses these models;
• Help to define and improve our internal standards for style, maintainability, and best practices for a high-scale web environment;
• Collaborate with other ML engineers and advise on the MLOps architecture from an infrastructure perspective;
• Respond to feature availability incidents and provide support for service engineers with customer incidents.

What you will bring

• Significant professional experience in Python backend infrastructure and Pytorch.
• Experience in working on scalability and maintainability challenges
• Production experience with Terraform, Kubernetes, Docker, and preferably GCP or equivalent technologies.
• Experience with NVIDIA Triton backend and post-production monitoring of large language models ( >10GB)
• High interest in defining infrastructure for large-scale ML recommendation engines 
• A genuine passion for learning as you will be solving the challenges of today, tomorrow, and many years to come.

About the team

Today, data scientists piece together data, tooling, and frameworks to get bespoke data science workloads running and producing business value. Data scientists then hand off models to engineering teams to attempt to deploy them to production. These teams speak different languages, use different tools, and have completely different workflows. This makes it very difficult to deploy data science workloads to production, increasing time to value and costs. One of our primary goals for our ModelOps stage is to reduce the complexities of data science workloads and integrate these workloads to easily be managed and developed within GitLab.

How GitLab will support you

• All remote, asynchronous work environment
• Unlimited PTO (paid time off)
• GitLab is home to many dedicated Team Member Resource Groups
• Benefits to manage your health, wealth, and well-being regardless of location
• Equity compensation & Employee Stock Purchase Plan offered
• A personal growth and development budget per year
• Parental leave: 16 weeks
• Home office support

#LI-MD5