Technical Support Engineer

Check out 30 new AI Systems Engineer opportunities posted on AI Chopping Block

The role involves designing and building execution environments for AI agents, including sandboxing, isolation, and reproducibility. It includes developing systems for agent orchestration across multi-step, tool-using workflows and building infrastructure for running, testing, and debugging code generated by models. Responsibilities also include creating state and memory systems that allow agents to persist context across long-running tasks, optimizing tokens, latency, reliability, and cost across Codex’s production fleet, and supporting model rollouts, capacity planning, and managing tradeoffs between quality, speed, and economics to maintain a fleet of frontier agents itionally, the job entails building shared platform capabilities that unblock product teams, partner teams, and open source Codex.

The Software Engineer in the Defence team will build and extend critical components of client deliverables across diverse software domains, deliver robust technical artefacts in both compiled and non-compiled languages, implement defined engineering patterns and practices tailored for the Defence sector, collaborate closely with Machine Learning Engineering and Data Science teams to integrate and refine technical solutions, apply rigorous software engineering best practices to enhance scalability and quality of codebases, and execute CI/CD processes while managing application deployments on Kubernetes and bare metal environments.

Lead a team of software engineers building detection and mitigation systems for frontier model misuse, focusing on model IP protection, distillation detection, and emerging risks from autonomous agents. Set the technical roadmap and execution strategy including prioritization, design, shipping, iteration, and impact measurement. Build production systems such as services, pipelines, tooling, instrumentation, and automation that can scale with frontier model usage.

Partner with Research and Product teams to translate evolving model capabilities into scalable tests, signals, and mitigations. Drive strong engineering fundamentals including architecture, reliability, monitoring, performance, and operational excellence. Hire and grow a team across backend, data systems, and applied ML engineering domains. Anticipate and address scalability challenges as agentic workflows advance.

The Staff Engineer is responsible for making hard architectural tradeoffs and owning the outcomes, such as choices between Durable Object SQLite and shared Postgres for session state, Cloudflare Workers CPU limits versus longer-running workloads, and single-tenant sandboxes versus multi-tenant pools. They design the system that handles concurrent agent sessions across integrations with consistent state. The role includes defining reliability and observability standards for the team, including SLOs, error budgets, tracing strategies, and incident response patterns.

The Staff Engineer reviews every significant pull request to set technical direction without blocking velocity and ships code daily, actively contributing architectural leadership alongside output. They work closely with the CTO on all architectural decisions that significantly impact the system.

Develop AI agents and software to automatically diagnose and repair hardware faults across massive NVIDIA and AMD GPU clusters. Create deep-level observability and diagnostic tools to monitor the health of high-density compute systems. Build testing suites using PyTorch and NCCL to ensure systems are ready for production. Develop automation for critical facility systems, including power management and advanced liquid cooling. Take full ownership of tools from initial code to deployment and operational support.

The Senior Systems Performance Engineer at Crusoe is responsible for leading the evaluation and establishment of New Product Introduction (NPI) across varied hardware architectures with a focus on Bare Metal and VM environments. They conduct deep-dive performance evaluations and workload characterizations across…