Important: This is a Revenue-Sharing Partnership Opportunity and not a fixed-salary position. If you are looking for a salaried role, this opportunity may not be suitable for you.
The Advantage
There is no earning cap. Your income is directly linked to your performance and successful closures.
...
You design computational engineering problems to challenge a frontier AI model. The problem must have an answer verifiable by code, and the problem has to require a specialized tool like OpenSeesPy, CalculiX, YADE, bempp-cl, or others. Generic numerical libraries on their own won't cut it. Each problem runs inside a sealed Linux container with the tool pre-installed and a programmatic judge that grades the model's answer. As an expert author, you: • Pick an anchor tool and design a problem that hinges on its solvers, simulation kernels, or domain-specific models. • Write a Python reference solution, supply input files and geometry definitions where needed. • Decide the numerical answer and how close the model needs to get — with a domain-appropriate tolerance — to count as right. • Test the problem against the model in batches of parallel attempts, tuning the problem difficulty until the agent only succeeds in a small number of attempts. • Once you're happy with the task, and it scores within range, the task goes to a senior reviewer in your subfield. They will provide feedback to ensure task quality is high. Calibration requires patience. You're tuning the problem against batches of parallel runs of the agent, aiming for a pass rate in the 10–30% band. Reaching that means rewriting load cases, tightening boundary conditions, and watching how the agents act. You'll learn how these agents cut corners, where a simulation stalls, where a solver converges. This time compounds in two directions. You come out of each task with deeper command of the anchor tool itself, and also get a hands-on working intuition for how a frontier model navigates complex structural and geotechnical problems.
What we look for
This opportunity is a good fit for engineers with an experience in python open to part-time, non-permanent projects. Ideally, contributors will have: • Degree in Civil Engineering or related field; • 2+ years of research, applied, or teaching experience; • Python proficiency for writing reference solutions; • Fluency with — or strong willingness to independently learn — at least one scriptable civil engineering package: OpenSeesPy, CalculiX, YADE, bempp-cl, or similar tools from the broader engineering catalogue; • Ability to design problems that genuinely require a specialized solver; • Strong written English (C1+). No prior experience with the listed tools? You're still welcome to apply — as long as you're ready to get up to speed on your own and hit the ground running.
...
You design computational engineering problems to challenge a frontier AI model. The problem must have an answer verifiable by code, and the problem has to require a specialized tool like Cantera, CoolProp, CalculiX, OpenFAST, or others. Generic numerical libraries on their own won't cut it. Each problem runs inside a sealed Linux container with the tool pre-installed and a programmatic judge that grades the model's answer. As an expert author, you: • Pick an anchor tool and design a problem that hinges on its solvers, simulation kernels, or domain-specific models. • Write a Python reference solution, supply input files and geometry or mechanism definitions where needed. • Decide the numerical answer and how close the model needs to get — with a domain-appropriate tolerance — to count as right. • Test the problem against the model in batches of parallel attempts, tuning the problem difficulty until the agent only succeeds in a small number of attempts. • Once you're happy with the task, and it scores within range, the task goes to a senior reviewer in your subfield. They will provide feedback to ensure task quality is high. Calibration requires patience. You're tuning the problem against batches of parallel runs of the agent, aiming for a pass rate in the 10–30% band. Reaching that means rewriting thermodynamic cycles, tightening material models and boundary conditions, and watching how the agents act. You'll learn how these agents cut corners, where a simulation stalls, where a solver converges. This time compounds in two directions. You come out of each task with deeper command of the anchor tool itself, and also get a hands-on working intuition for how a frontier model navigates complex thermal, structural, and fluid mechanics problems.
What we look for
This opportunity is a good fit for engineers with an experience in python open to part-time, non-permanent projects. Ideally, contributors will have: • Degree in Mechanical Engineering or related field; • 2+ years of research, applied, or teaching experience; • Python proficiency for writing reference solutions; • Fluency with — or strong willingness to independently learn — at least one scriptable mechanical engineering package: Cantera, CoolProp, CalculiX, OpenFAST, YADE, or similar tools from the broader engineering catalogue; • Ability to design problems that genuinely require a specialized solver; • Strong written English (C1+). No prior experience with the listed tools? You're still welcome to apply — as long as you're ready to get up to speed on your own and hit the ground running.
...
Test the problem against the model in batches of parallel attempts, tuning the problem difficulty until the agent only succeeds in a small number of attempts.
Once you're happy with the task, and it scores within range, the task goes to a senior reviewer in your subfield. They will provide feedback to ensure task quality is high.
Use evidence-based practices to address specific developmental areas such as communication, motor skills, social interaction, and cognitive development.
Offer guidance and training to parents and caregivers to support the child’s learning and development outside of sessions.
Keep detailed records of each child’s progress, adjusting the intervention approach as needed to meet evolving needs.
...
Test the problem against the model in batches of parallel attempts, tuning the problem difficulty until the agent only succeeds in a small number of attempts
Once you're happy with the task, and it scores within range, the task goes to a senior reviewer in your subfield. They will provide feedback to ensure task quality is high
Test the problem against the model in batches of parallel attempts, tuning the problem difficulty until the agent only succeeds in a small number of attempts
Once you're happy with the task, and it scores within range, the task goes to a senior reviewer in your subfield. They will provide feedback to ensure task quality is high