Forward Deployed Inference Engineer

About the role

<h2>About Tensordyne</h2> <p>Tensordyne is building a new class of AI inference system designed for high-performance, power-efficient deployment of the world’s most demanding generative AI workloads.</p> <p>Our platform combines purpose-built silicon, new AI math, optimized scale-up networking, and memory architecture into a tightly integrated system purpose built for large-scale AI inference. We work with hyperscalers, Neoclouds, frontier model developers, enterprises, and infrastructure partners operating at the leading edge of AI.</p> <p>As Tensordyne moves from system development into silicon bring-up, customer validation, beta deployments, and production rollout, we are building the technical customer organization that will sit directly between our engineering teams and the companies deploying the platform.</p> <h2>Role summary</h2> <p>We are looking for a <strong>Forward Deployed Inference Engineer</strong> who combines deep AI systems expertise with strong customer instincts. This person will own the path from a customer workload or model request to a technical result and, where needed, to an optimized model running successfully on Tensordyne hardware and software.</p> <p>The role sits at the intersection of model architecture, inference performance, systems optimization, developer tooling, and customer deployment. You will work hands-on with engineering while also acting as a technical bridge to Product, BizDev, Sales, and customers.</p> <h2>What you will do</h2> <ul> <li>Turn customer workloads into fast, credible performance answers through <strong>profiling &amp; benchmarking</strong>. Define relevant KPIs, compare against competitive baselines, and keep our evaluation methodology current with external benchmarks.</li> <li><strong>Model enablement &amp; optimization:</strong> convert and bring up customer models on the Tensordyne stack, validate numerical quality, identify performance bottlenecks, and work with compiler, runtime, kernel, and system teams to improve results.</li> <li><strong>Deployment / forward engineering:</strong> work directly with customers and partners on technical PoCs, integration, deployment, and debugging; translate requirements into measurable acceptance criteria for quality, latency, throughput, and other relevant KPIs.</li> <li>Track <strong>profiling-to-hardware accuracy</strong> by continuously comparing profiling/simulation results with actual hardware deployments, explain material gaps, and flag missing capabilities in the compiler, SDK, inference server, KV-cache management, or adjacent systems to the owning teams.</li> <li>Turn repeated customer-specific learnings into reusable tooling, documentation, benchmarks, or product improvements.</li> </ul> <h2>Core qualifications</h2> <ul> <li>Strong hands-on experience with <strong>AI models and inference systems</strong>, especially dense and MoE LLMs (Llama, DeepSeek, Qwen, GPT-OSS, Kimi, GLM), and VLM, speech and diffusion models.</li> <li>Strong <strong>Python and PyTorch</strong> skills and the ability to understand and modify model code.</li> <li>Experience <strong>profiling, benchmarking, or optimizing model inference</strong> and reasoning about latency, throughput, memory, and utilization.</li> <li>Strong problem-solving and communication skills, with the ability to drive ambiguous technical problems across team boundaries.</li> <li>Proficiency in using AI-powered developer tools (e.g., Claude Code, Cursor).</li> </ul> <h2>Strong pluses</h2> <ul> <li>Experience with <strong>LLM serving and deployment stacks</strong> such as vLLM, SGLang, or similar systems.</li> <li>Experience working directly with <strong>customers or external technical partners</strong>.</li> <li>Experience bringing models up on <strong>new accelerators or non-standard hardware</strong>, including performance debugging across framework/runtime/hardware boundaries.</li> <li>Practical experience with production inference techniques or environments such as <strong

Skills