As a Senior Systems Engineer – AI/ML Infrastructure at Deepgram, your primary responsibilities include designing and maintaining large-scale distributed systems infrastructure, focusing on network architecture, storage solutions, and compute platforms essential for AI/ML workloads. You'll manage bare-metal GPU clusters, implement comprehensive monitoring and automation solutions, and ensure high performance and reliability across the infrastructure.

To qualify for the Senior Systems Engineer – AI/ML Infrastructure position at Deepgram, you'll need a minimum of 5 years of experience in infrastructure engineering or similar roles. A strong background in network engineering, managing large-scale storage systems, and proficiency with container orchestration platforms like Kubernetes and Slurm are critical. Additionally, experience with GPU infrastructure management and strong automation skills will set you apart.

In the role of Senior Systems Engineer – AI/ML Infrastructure at Deepgram, you will work with cutting-edge technologies including bare-metal GPU compute clusters, Kubernetes for container orchestration, and various large-scale storage solutions. You'll also engage with automation tools for infrastructure management and focus on optimizing networking for distributed systems.

Deepgram is committed to the professional development of its team members, including those in the Senior Systems Engineer – AI/ML Infrastructure role. With opportunities to work on advanced projects, access to the latest technologies, and a collaborative environment, you’ll have many avenues to learn and advance your skills in AI and infrastructure engineering.

At Deepgram, we're proud of our inclusive and collaborative culture, which is vital for roles like the Senior Systems Engineer – AI/ML Infrastructure. We value diverse perspectives, encourage teamwork, and focus on doing what's right for our customers. This connection fosters innovation and growth, helping you thrive in your role.

When addressing your experience with GPU infrastructure, focus on specific projects where you managed bare-metal GPU clusters for AI training and inference. Discuss the challenges you faced, the optimizations you implemented, and the results achieved. Highlight any performance metrics that demonstrate your effectiveness.

For reliability in distributed systems, discuss methods you have implemented, such as robust monitoring strategies, alerting systems, and redundancy measures. Emphasize your ability to design architectures that mitigate failures and ensure seamless operation under various loads.

You should mention your experience with platforms like Kubernetes and Slurm, providing examples of how you've used them to deploy and manage applications. Explain your role in optimizing resource allocation and scaling applications efficiently in a production environment.

Think of a specific complex networking issue you've faced and provide details on the technical aspects. Discuss the strategies you deployed, the tools you used, and the overall impact of your solution. This showcases your critical thinking and problem-solving abilities.

For automation, discuss the tools and scripts you've implemented to streamline infrastructure management tasks, such as deploying resources, backup operations, or monitoring. Explain how your automation efforts improved efficiency and reduced downtime, emphasizing your scripting skills.

Provide examples of large-scale storage systems you've architected or maintained, such as distributed file systems or caching solutions. Highlight your understanding of data redundancy, retrieval speeds, and backup strategies, so interviewers can see your expertise in handling massive volumes of data.

Discuss specific AI/ML projects you've been involved in, detailing your role in shaping the infrastructure to support these workloads. Mention how your experience with GPU optimization has enhanced performance and contributed to successful outcomes in machine learning tasks.

Talk about your experience with budgeting and capacity planning for infrastructure projects, providing tangible examples. Discuss strategies you've implemented to minimize costs while maintaining performance, such as using cost-effective storage solutions or optimizing resource utilization.

Explain how you assess current usage patterns and future growth needs. Discuss any tools or methodologies you utilize for monitoring capacity and predicting necessary adjustments, ensuring you're prepared for changing demands.

Define success in terms of performance metrics such as system uptime, efficiency gains from optimizations, successful project completions, and positive feedback from stakeholders. This showcases your understanding of key performance indicators that align with Deepgram’s goals.