Machine Learning Infrastructure Tech Lead

About the role

Reducto is seeking a hands‑on ML Infrastructure Tech Lead to own and evolve the systems that enable high‑performance model training and inference. You will split your time between building core infrastructure and setting the technical roadmap for the ML platform.

Key responsibilities

• Design and maintain the end‑to‑end training and inference stack, from low‑level kernels to Kubernetes orchestration.
• Optimize GPU utilization, latency, throughput and cost across multi‑node, multi‑GPU workloads.
• Develop benchmarks, identify bottlenecks and drive infrastructure investments.
• Partner with ML and Platform engineers on architecture, capacity planning and technical prioritisation.
• Raise the engineering bar through design reviews, mentorship and hands‑on leadership.

Required profile

• 5+ years building production infrastructure with significant ML systems experience.
• Proven ability to lead ambiguous, high‑impact projects from concept to production.
• Strong Python and systems‑engineering skills, comfortable with Kubernetes and distributed training/serving frameworks.
• High standards for quality, reliability and operational excellence in a fast‑growing startup.

Required skills

• Python
• Kubernetes
• GPU training/inference performance optimisation
• CUDA, Triton or custom model‑serving kernels (preferred)
• Experience with frameworks such as vLLM, SGLang, PyTorch, TensorRT‑LLM, Ray (preferred)

What we offer

• Unlimited PTO
• Free daily lunch
• Commuter reimbursement
• Comprehensive medical, dental and vision insurance
• Health & wellness budget up to $150/month
• Parental leave and flexible scheduling