Forward Deployed Engineer

📋 Description

• • Own the full lifecycle of high-impact AI deployments. You will embed on-site with Fortune-500 customers in manufacturing, energy, and logistics, translating messy, real-world operations into elegant, production-grade systems that run 24/7.
• • Architect end-to-end solutions that fuse Generative AI, Computer Vision, and classical ML with legacy SCADA, MES, and ERP stacks. Expect to design everything from GPU-accelerated vision pipelines that read analog gauges to LLM agents that reason over work-order histories.
• • Rapidly prototype in Python, TypeScript, or Go, then harden code into fault-tolerant micro-services running on Kubernetes in customer VPCs. You’ll write Dockerfiles, Terraform, Helm charts, and GitHub Actions that survive audits and red-team reviews.
• • Lead customer workshops to map pain points, define success metrics, and secure stakeholder buy-in. You’ll whiteboard architectures at 9 a.m., pair-program with plant engineers at noon, and present ROI dashboards to C-suite executives at 5 p.m.
• • Build and tune ML pipelines: ingest terabytes of video and sensor data, craft prompt templates, fine-tune open-source LLMs, and deploy retrieval-augmented generation (RAG) systems that answer technical questions in <500 ms.
• • Optimize models for latency, throughput, and cost. You’ll profile GPU kernels, quantize weights, and implement adaptive batching so that a single A100 can serve 200+ concurrent inspection requests.
• • Integrate with existing plant networks, historians, and alarm systems. You’ll write OPC-UA connectors, Modbus gateways, and RESTful APIs that push AI-derived insights back into the control room in real time.
• • Troubleshoot across the stack: from camera calibration drift to Kubernetes DNS hiccups to subtle prompt hallucinations. You’ll carry a ruggedized laptop onto the shop floor, SSH into edge boxes, and ship fixes before the next shift starts.
• • Create reusable accelerators—SDKs, Terraform modules, and Jupyter-based notebooks—that internal teams and customers can leverage for future projects, cutting deployment time from months to days.
• • Mentor junior Forward Deployed Engineers and Solutions Architects. You’ll run brown-bag sessions on vision transformers, review pull requests, and codify best practices that scale across a growing customer base.
• • Influence the Nexxa product roadmap by distilling field feedback into crisp PRDs and prioritizing features that unlock new verticals. Your GitHub issues today become the platform capabilities of tomorrow.
• • Travel up to 50 %—sometimes on short notice—to customer sites across North America. You’ll spend Monday debugging in a refinery, Wednesday sprint-planning in Sunnyvale, and Friday celebrating a successful go-live with plant operators.