Engineering Expert (AI Evaluation, Simulation & Design)
- Company: Turing
- Category: AI / Machine Learning
- Location: Remote
- Rate: Rate not specified
- Experience: Senior Level
- Posted: November 16, 2026
About the Role
Turing is hiring experienced AI Evaluation Engineers in engineering simulation and design, under the title Engineering Expert. You write and validate "model-breaking", simulation-based engineering design problems used to train and evaluate state-of-the-art AI agents. The work spans the major engineering disciplines, including Electrical, Mechanical, Control Systems, Aerospace, Systems and Robotics. You create complex, multi-constraint tasks in which AI agents must interpret requirements, weigh trade-offs, configure open-source simulation tools, diagnose failures and iterate toward valid solutions. You analyse agent execution logs, expose systematic reasoning gaps and build automated, objective graders. The work is remote.
Job responsibilities
- Model-breaking problem design: write original, self-contained engineering design tasks with competing constraints, explicit optimisation targets, validated reference solutions and objective autograders
- Environment and simulation integration: build, run and validate problem environments using open-source simulation tools and custom Python test benches
- Trajectory analysis and failure-mode taxonomy: evaluate coding-agent outputs and execution logs across repeated trials to find systematic failure modes, such as misreading simulator feedback, converging on a design too early or producing physically impossible geometries
- Difficulty calibration and benchmark refinement: refine problem difficulty based on measured model performance, without introducing ambiguity or missing information
- Cross-functional collaboration: work with AI researchers, pod leads and domain experts to bring high-rigour benchmarks into the model evaluation pipeline
Domains
- Electrical Engineering
- Mechanical Engineering
- Aerospace Engineering
Offer details
- Commitment: up to 40 hours per week, with 4 hours of overlap with Pacific Time (PST)
- Engagement type: contractor
- Engagement length: up to 12 weeks
Evaluation process
- Candidates submit a job interest form and complete an assessment
- People who pass the assessment go on to a delivery interview and proceed from there
About Turing
- Based in San Francisco, California, Turing describes itself as a research accelerator for frontier AI labs and a partner for enterprises deploying advanced AI systems
- It supports AI labs with high-quality data, training pipelines and researchers who specialise in coding, reasoning, STEM, multilingual and multimodal work, and agents
- It also helps enterprises turn AI proofs of concept into dependable systems with measurable business results
Job details
- Employer: Turing
- Location: Remote
- Work type: Contract, up to 40 hours per week for up to 12 weeks
- Field: Engineering
- Required skills listed: Classical Mechanics, Electrical Networks
- Pay: not published on the listing
Requirements
- Education and expertise: a relevant degree in Electrical, Mechanical or Aerospace engineering, with 5 or more years of hands-on engineering design experience
- A Bachelor's or higher degree in the relevant engineering field is a must
- Simulation tooling: proficient with at least one relevant open-source simulation package (for example ngspice, PySpice, OpenFOAM, FEniCSx, CalculiX, python-control, CadQuery, build123d, OpenModelica, Cantera or Gmsh), plus strong Python scripting
- AI evaluation and failure diagnostics: hands-on experience with modern LLMs and coding agents and with evaluation concepts (pass@k, failure-mode analysis, nondeterministic behaviour), able to audit trajectory logs and isolate reasoning and tool-use failures
- Required: experience in AI evaluation, data annotation, content review, quality assurance or a related analytical role
- Domain rigour and precision: uncompromising attention to physical plausibility, unit consistency, boundary conditions, convergence criteria and technical documentation
- Availability: weekend on-call availability is required (a part-time engagement is acceptable)
- Technical setup: a personal desktop or laptop with a stable, high-speed internet connection
- Able to overlap 4 hours a day with Pacific Time