[5X9IHLIM] LLM Engineering Expert

Hello,

Greetings from ZettaMine!!!

Freelancing Opportunity :- We are currently having multiple opportunities for LLM – Engineering Experts to support an AI-focused engineering evaluation project across Bangladesh, India, Indonesia, Egypt, Ghana, Kenya, Nigeria, Turkey, Brazil, Colombia, and approved states in the USA.

Role: LLM – Engineering Expert

Experience: 10+ years of hands-on engineering design experience

Mode: Remote

Engagement: Short-Term Contract / Freelance

Start: Immediately

Mandatory Eligibility Criteria

Technical & Professional Qualification:

• Master’s degree or PhD in Electrical Engineering, Mechanical Engineering, Aerospace Engineering, or a related engineering discipline.
• 10+ years of hands-on engineering design experience.
• Strong proficiency in at least one domain-relevant open-source simulation package such as ngspice, PySpice, OpenFOAM, FEniCSx, CalculiX, python-control, CadQuery, build123d, OpenModelica, Cantera, or Gmsh.
• Strong Python scripting skills for engineering simulations, automation, analysis, and test benches.
• Hands-on experience with modern LLMs and coding agents, along with knowledge of AI evaluation concepts such as pass@k, failure-mode analysis, and nondeterministic behavior.
• Ability to audit AI agent trajectory logs, identify reasoning and tool-use failures, and diagnose systematic model behavior.
• Strong understanding of physical plausibility, unit consistency, boundary conditions, convergence criteria, and engineering documentation.
• Strong analytical and problem-solving skills with the ability to design complex engineering problems and evaluate technical solutions.
• Experience working with engineering simulation, computational modeling, design validation, or related technical environments.

What You’ll Work On:

• Author original, self-contained engineering design and simulation problems designed to challenge and evaluate advanced AI agents.
• Develop problems with competing constraints, explicit optimization objectives, validated reference solutions, and objective automated graders.
• Build, run, and validate problem environments using open-source simulation tools and custom Python test benches.
• Evaluate AI-generated engineering solutions and coding-agent execution logs across repeated trials.
• Identify systematic AI failure modes, including misinterpretation of simulator feedback, premature design convergence, physically impossible geometries, and incorrect tool usage.
• Analyze model performance and iteratively calibrate benchmark difficulty without introducing ambiguity or missing information.
• Translate complex engineering requirements into practical, measurable AI evaluation tasks.
• Validate simulation outputs for physical correctness, engineering feasibility, unit consistency, boundary conditions, and convergence.
• Collaborate with AI researchers, pod leads, and other engineering domain experts to integrate high-rigor benchmarks into AI model evaluation pipelines.
• Develop reusable evaluation assets, reference solutions, test benches, simulation environments, and automated grading systems.
• Contribute technical insights to improve the performance and reliability of frontier AI models.

Availability:

• Minimum 40 hours per week.
• Minimum 4 hours of overlap with PST.
• Weekend on-call availability is required.
• Immediate availability is preferred.
• Part-time engagement may be considered based on project requirements.

Technical Infrastructure:

• Personal desktop/laptop suitable for engineering simulation and development work.
• Stable, high-speed internet connection.
• Ability to work independently in a remote environment.

Others:

• LinkedIn profile is mandatory for review. Please ensure the LinkedIn profile link is included in the resume.
• Strong written and verbal communication skills.
• Ability to work independently and collaborate effectively with remote global teams.

Interested candidates: Please share your updated CV with LinkedIn profile link at: raviteja.k@zettamine.com

Thanks & Regards,

RaviTeja K

ZettaMine