Reducto Machine Learning Infra Engineer at Reducto building and maintaining training and inference frameworks for scalable ML. Collaborate with ML and Platform teams to improve GPU utilization and system observability.
Responsibilities
As an ML Infra Engineer, you’ll play a key role in building the inference and training frameworks that make it possible to deliver results at scale. You’ll collaborate closely with our ML and Platform teams to scale training across nodes, develop faster and more efficient serving, and create observability across the stack. This is a high-impact role where you’ll help define what high performance ML training and inference look like at Reducto.
Build, and maintain our training and inference stack with an emphasis for fast iteration on training + flexibility for exploring new methods and high performance in inference.
Develop benchmarks for both sets of stacks to identify bottlenecks.
Explore SOTA advances in training and inference and work to apply them.
Design systems for scaling model training across multi-node, multi-GPU environments with strong reliability and observability.
Scale distributed training and inference workloads across large GPU clusters while improving utilization, reliability, and cost efficiency.
Build the tooling, abstractions, and observability that help ML engineers move faster from experiment to production.
You’ll Thrive Here If You:
Hold yourself to a high bar for quality and precision.
Enjoy solving complex problems and building from first principles.
Have strong Python skills + a background in systems engineering.
Are comfortable with Kubernetes and distributed training frameworks.
Qualification
3+ years of experienceMore about Reducto
Required
3+ years of experience
Strong Python skills and background in systems engineering
Comfortable with Kubernetes and distributed training frameworks
Preferred
Have experience at an early-stage or high-growth startup.
Have developed in open source training/inference stacks in a meaningful way.
Are excited to set up distributed inference across 100s-1000s of GPUs.
Care deeply about combining technical excellence with business impact.
This is an in person role at our office in SF. We’re an early stage company which means that the role requires working hard and moving quickly. Please only apply if that excites you.
More about Reducto
Nearly 80% of enterprise data is in unstructured formats like PDFs
PDFs are the status quo for enterprise knowledge in nearly every industry. Insurance claims, financial statements, invoices, and health records are all stored in a structure that’s simply impractical for use in digital workflows. This isn’t an inconvenience—it’s a critical bottleneck that leads to dozens of wasted hours every week.
