You are viewing a preview of this job. Log in or register to view more details about this job.

Automation and AI Agent Engineer

Scrap Labs
Email

AUTOMATION AND AI AGENT BUILDING ENGINEER

Boulder, CO (On-site or Hybrid)

Scrap Labs was founded under the belief that high-performance metal 3D printing should be accessible to builders, labs, and small manufacturers, not just massive corporations. We build ultra-low-cost metal additive manufacturing systems and the software that makes them easy to use. Our goal is to turn advanced manufacturing into a push-button workflow through automation, tight feedback loops, and relentless iteration.

AUTOMATION AND AI AGENT BUILDING ENGINEER

As a member of the Scrap Labs software team, you will design, build, and deploy AI-powered agents that automate real operational work across engineering, manufacturing, and customer workflows. You will ship systems that reason over data, call tools, take actions, and close loops with measurements from the physical world. This role is hands-on, fast-paced, and focused on delivering working product, not demos.

RESPONSIBILITIES

  • Build agentic systems that automate workflows such as procurement, inventory management, administrative tasks, quoting, print planning, job launch, monitoring, maintenance guidance, and customer support
  • Implement tool use: API integrations, database operations, file system actions, web and UI automation, and human-in-the-loop approval where needed
  • Build computer use agents (CUA) that can operate real software interfaces to get work done: browser apps, internal dashboards, vendor portals, and desktop tools
  • Develop robust GUI automation stacks (Playwright, Selenium, VNC, remote desktops) with anti-flake strategies, step verification, and audit logs
  • Develop retrieval and knowledge systems (RAG) over internal documentation, CAD metadata, logs, tickets, and manufacturing process data
  • Own evaluation and reliability: offline test sets, regression harnesses, agent scoring, guardrails, and failure analysis
  • Deploy and operate services in production: packaging, observability, rate limiting, logging, and secure secrets management
  • Build security into agent systems: least-privilege tool access, authentication/authorization, secrets hygiene, audit trails, and safe data handling
  • Defend against agent-specific threats (prompt injection, data exfiltration, tool misuse) with guardrails, sandboxing, and human approvals where appropriate
  • Optimize latency and cost: model selection, caching, batching, quantization, and local inference performance
  • Run and tune local models on real hardware: multi-GPU servers, model serving, GPU scheduling, and inference performance profiling
  • Collaborate with hardware, controls, and manufacturing teams to define interfaces and close feedback loops from sensors and process outcomes
  • Maintain high engineering standards: clean code, code reviews, versioning, and documentation that stays accurate

BASIC QUALIFICATIONS

  • Bachelor’s degree in Computer Science, Engineering, or related technical field, or equivalent practical experience
  • Strong proficiency in Python and at least one additional language used for production services (TypeScript/JavaScript, Go, or similar)
  • Experience building and shipping software systems end-to-end (internship, research, open-source, or professional)
  • Comfort in a Linux environment with Git, containers (Docker), and service deployment
  • Ability to move quickly, prioritize ruthlessly, and deliver working systems under uncertainty

PREFERRED SKILLS AND EXPERIENCE

  • Experience building LLM applications, agents, or automation systems (tool calling, function schemas, planning, memory, orchestration)
  • Experience working alongside coding agents (Claude Code, Codex, OpenCode, or similar) to accelerate development, debugging, and code review
  • Experience with computer use agents (CUA) or RPA-style systems that take actions in GUIs, including step verification and safe fallback behavior
  • Strong experience with web automation (Playwright or Selenium) and a practical approach to building resilient, non-brittle automations
  • Familiarity with common LLM frameworks and patterns (LangChain, LlamaIndex, OpenAI-style APIs, structured outputs)
  • Experience running models locally (vLLM, SGLang, Triton, CUDA basics) and tuning inference for throughput and latency
  • Comfort with GPU hardware and systems: multi-GPU configurations, CUDA basics, NCCL fundamentals, and performance debugging on Linux
  • Experience with quantization and deployment tradeoffs (FP16, BF16, INT8, 4-bit), plus profiling and regression testing around quality
  • Experience with retrieval systems (vector databases, embeddings, chunking strategies, reranking) and search evaluation
  • Practical knowledge of software observability (metrics, tracing, dashboards, alerting) and incident-style debugging
  • Experience with application security and secure-by-default design (RBAC, OAuth/OIDC, API key management, OWASP basics)
  • Experience with threat modeling and security reviews for systems that take actions
  • Familiarity with infra and container security (network segmentation, Docker hardening, least privilege)
  • Experience with workflow systems and integrations (ERP, ticketing, Slack/Discord bots, Google Drive/Nextcloud, email)
  • Experience with web automation and UI agents, including robust selectors and anti-flake strategies
  • Interest in manufacturing, robotics, 3D printing, and hardware-software integration

ADDITIONAL REQUIREMENTS

  • Ability to work on-site in Boulder, CO at least 2-3 days per week (adjustable based on candidate and project needs)
  • Work schedule may vary based on integration pushes and operational needs, and flexibility is required

COMPENSATION AND BENEFITS

  • Base salary: $80k - $120k depending on experience
  • Additional compensation: performance bonus or revenue share tied to company outcomes
  • Onsite free weight workout equipment
  • Health: medical, dental, and vision insurance (company contribution)
  • Time off: paid time off, paid holidays, and paid sick leave in accordance with Colorado law
  • Learning: annual learning stipend for role-relevant training and conferences

EQUAL OPPORTUNITY

Scrap Labs is an Equal Opportunity Employer. Employment decisions are based on merit, competence, and qualifications and will not be influenced by any legally protected status.

TO APPLY

Send a resume and links to 1 to 3 examples of shipped work (GitHub, project write-ups, demos, or papers). Include a short note describing a system you built that made a process faster, cheaper, or more reliable.