Member of Technical Staff

Summary

Join Outspeed, a company building real-time AI infrastructure, as an early Member of Technical Staff. You will work on various aspects of the platform, including core inference engines, orchestration services, and customer-facing tools. The role involves rapid prototyping, production system hardening, and shipping code used globally. You will own end-to-end feature work, optimize GPU inference pipelines, and collaborate with customers to resolve issues. You will also contribute to engineering culture by mentoring and shaping the technical roadmap. This position is ideal for engineers who enjoy diverse tasks and want significant impact.

Requirements

• 2+ years of professional software engineering, with deep proficiency in at least one systems language (Go/Rust/C++) and one high-level language (Python/TypeScript)
• Proven track record shipping production-grade distributed systems —you’ve designed, implemented, and operated services that run at scale and stay up
• Hands-on experience with end-to-end ML workflows : data pipelines, model training/tuning, packaging, and low-latency serving (PyTorch + CUDA or similar)
• Strong product sense and the ability to translate ambiguous problems into well-scoped engineering work
• Excellent written & verbal communication; comfortable leading design docs and code reviews in a fast-moving, asynchronous environment

Responsibilities

• Own end-to-end feature work—from RFC and design review to infrastructure-as-code, monitoring, and post-launch iteration
• Optimize GPU inference pipelines, extending our TypeScript/React console, and designing a new audio-streaming protocols
• Pair with customers to debug latency spikes, then upstreaming the fix into our orchestration layer
• Influence engineering culture: instituting best practices, mentoring newer hires, and shaping our technical roadmap alongside the founders

Preferred Qualifications

• Expertise running containerized workloads with Docker, Kubernetes, or Nomad , plus IaC with Terraform or Pulumi
• Knowledge of public-cloud primitives (SDN, block storage, secrets/identity, spot fleet management) and how to squeeze every dollar out of them
• Familiarity with modern inference tooling (e.g., vLLM, Ray, Triton, llama.cpp) and a solid mental model of GPU/TPU performance bottlenecks
• A history of thriving in early-stage startups: rapid iteration, “build → measure → learn” loops, and wearing multiple hats at once

Benefits

• Competitive salary + Equity
• Health, dental, and vision insurance
• Bonuses based on performance