The company is building software that lets SI agents and multimodal models work with real equipment through an existing hardware integration stack.
The work centers on applied SI and perception workflows that turn operational data and equipment interfaces into safe, usable SI-assisted systems.
The role sits in the Material Engineering Division within the Engineering Directorate.
The job is hybrid, mixing in-person and remote days, with the option to work from home one or more days a week.
The role may be filled at either level, based on knowledge and related experience as judged by the hiring team. Higher-level duties apply if you are hired at that level.
Responsibilities
Build agent-facing tool definitions and task workflows that call documented equipment interfaces and platform services.
Create multimodal perception and decision-support workflows that use camera, telemetry, and process data.
Refine prompt logic, tool use, evaluation methods, and workflow reliability for SI applications in the physical world.
Turn real operational tasks into SI-assisted workflows that are practical, testable, and easy to maintain.
Work with embedded, platform, and deployment engineers in a controlled environment to set safe action limits, approval points run by a person or system, and fallback behaviors, using established external safety and security systems.
Test and debug applied SI workflows through thorough logging, repeated refinement, and disciplined engineering practice.
Take on other tasks as assigned.
Additional responsibilities at the SES.2 level:
Design and assess agent workflows that join perception, tool use, and equipment actions across several system components, input types, and operational steps.
Improve reusable evaluation harnesses, guardrails, and failure analysis for SI-assisted workflows, and use the results to inform technical decisions.
Work with embedded and SI platform engineers to turn device capabilities, operational needs, and evaluation results into reusable agent workflows, and give informal guidance to less experienced contributors.
Requirements
You must be able to obtain and keep a U.S. DOE Q-level security clearance, which requires U.S. citizenship.
A bachelor's degree in computer science, software engineering, robotics, or a related field, or an equivalent mix of education and related experience.
Proficiency in Python or a similar programming language for building applications, prototypes, and internal tools with software and systems engineers.
Experience from coursework, projects, research, internships, or paid work building LLM, multimodal, agent-based, or tool-calling applications.
Experience designing, building, testing, and debugging workflows where models use APIs, services, data sources, or operational systems, including structured tool or API contracts such as JSON Schema or OpenAPI, error handling, logging, and evaluation.
Experience working with real-world data from sensors, cameras, devices, or operational systems.
Familiarity with sound software practices such as version control, automated testing, and peer code review.
Additional qualifications at the SES.2 level:
Ability to independently build and evaluate multicomponent agent, multimodal, or tool-calling workflows for operational use.
Ability to scope and debug moderately complex workflow failures that involve perception, tool use, system integration, and operational constraints.
Exposure to robotics workflows, ROS, or ROS 2.
Experience with vision-language models, vision-language-action models, or multimodal perception pipelines.
Familiarity with simulation, digital twins, or hardware-in-the-loop testing.
Familiarity with workflow orchestration, agent tracing or evaluation systems, prompt or configuration versioning, replay-based testing, or event and telemetry systems such as MQTT or NATS.
Conditions
Pay ranges from $121,830 to $185,544 annually across levels.
The SES.1 level pays $121,830 to $154,500 annually.
The SES.3 level pays $146,340 to $185,544 annually.
This range is the lowest to highest the company in good faith expects to pay for this role when posted. Pay will never fall below any applicable local minimum wage. Placement within the range depends on factors such as specific competencies, education, qualifications, certifications, experience, skills, seniority, geographic location, performance, and business or organizational needs.
This is a career indefinite position, open to internal lab employees and external candidates.
Schedules can be flexible, depending on project needs.
The position requires a DOE Q-level clearance. If selected, you will undergo a federal background investigation to confirm eligibility for access to classified information or matter. All L or Q cleared employees are subject to random drug testing. Q-level clearance requires U.S. citizenship.
External candidates selected must pass a post-offer, pre-employment drug test, which includes testing for marijuana because federal law applies as the company is a federal contractor.
Some roles require working in Limited Areas, where personal and laboratory mobile devices, including phones, tablets, fitness trackers, wireless headphones, and other Bluetooth or wireless devices, are not allowed. Medical devices that pair with a mobile device must follow the same rules. Sensitive Compartmented Information Facilities need separate approval. Hearing aids without wireless capability, or with wireless turned off, are allowed in Limited Areas, Secure Space, and Transit or Buffer Space within buildings.
Be cautious of recruitment scams using the company's name. The company never makes an offer without a personal interview and never charges a fee to join. Current openings appear on the careers page. Do not respond to suspicious offers.