Technical Problem Author (Multi-Domain)

What you would do

• Author complex technical problems in your specialty, such as aerodynamics, mechanical design, or circuit design, with appropriate rigor and scope
• Evaluate AI-generated responses for technical correctness, practical applicability, and engineering soundness
• Apply practical engineering judgment to real-world scenarios, design choices, and failure analysis cases alongside theoretical problems
• Identify and resolve ambiguities in problem statements, proposing clarifications that improve technical accuracy
• Provide structured feedback and written explanations that enhance AI learning and reasoning capabilities

Who they want

• MS or PhD in a relevant engineering field, or 5-8+ years of substantial industry experience in your specialty
• At least 5 years applying technical expertise within your specialty including structural analysis, computational fluid dynamics, circuit systems, or equivalent
• Proven track record authoring technical problems for olympiads, licensure exams, academic qualifiers, standards committees, or competitions
• Publications, patents, recognized achievements in your field, with demonstrated technical rigor through achievements in competitions or professional contributions
• Professional-level English and exceptional written and verbal communication skills; experience with industry-standard tools in your specialty

Main skills

What the interview asks about

1. 1.Technical problem authoring with appropriate scope

   Problems must be difficult enough to test genuine reasoning but clear enough to have a defensible answer - this balance is learned through experience setting problems for high-stakes contexts.

   For example: “Write me one aerodynamics problem that a PhD candidate should solve in 45 minutes but a junior engineer might struggle with. Include the key insight needed and explain where AI might miss it.”

2. 2.AI response evaluation and criticism

   Judging AI technical responses requires distinguishing between correct answers, plausible-sounding wrong answers, and responses that make unrealistic assumptions about materials or real-world constraints.

   For example: “An AI proposes a mechanical design solution that is mathematically sound but uses a material not available in that cost range for this application. How would you explain this error and what feedback would help the model improve?”

3. 3.Domain troubleshooting and ambiguity resolution

   Real engineering problems often contain vague specs or conflicting constraints - recognizing and clarifying these separates experts from generalists in your field.

   For example: “A circuit design problem you created is producing unexpectedly varied AI responses. Walk me through how you would diagnose whether this is ambiguous problem wording, unrealistic design assumptions, or legitimate alternative solutions.”

4. 4.Practical versus theoretical engineering judgment

   The difference between theoretical correctness and real-world feasibility is where deep domain expertise becomes visible, and where AI often struggles most.

   For example: “Create two versions of a mechanical design problem: one purely theoretical with ideal conditions, one with practical constraints like off-the-shelf component availability or manufacturing tolerances. Explain what each version tests.”

5. 5.Structured feedback for AI model training

   Helping AI systems improve requires explaining not just what is wrong but why it is wrong and what correct reasoning looks like in your specialty.

   For example: “Provide feedback on this AI response to an aerodynamics design challenge that gets the answer right but uses imprecise terminology and makes an unstated assumption. Write feedback that would help the model reason more rigorously next time.”

A task you may get

Create an advanced technical problem in your domain and provide three AI responses: correct with good reasoning, correct with flawed logic, and incorrect. Write feedback for each response.

How to prepare

• Review problems from olympiads, licensure exams, or technical competitions in your field to understand difficulty calibration
• Examine how ambiguous problem statements can produce multiple valid interpretations and how experts disambiguate
• Practice evaluating technical explanations at varying levels of rigor and identifying where engineering shortcuts miss important details
• Collect examples of unrealistic assumptions in technical problems you have encountered and how experts would flag them

The facts

Pay

$60–110/hr

Open to

Bangladesh, Hong Kong, India, Indonesia, Japan, Kazakhstan, Kyrgyzstan, Malaysia, Pakistan, Philippines, Singapore, Sri Lanka, Taiwan, Thailand, Uzbekistan, Vietnam, Austria, Belarus, Belgium, Denmark, France, Germany, Greece, Italy, Netherlands, Portugal, Russia, Spain, Switzerland, United Kingdom, Argentina, Brazil, Chile, Colombia, Mexico, Peru, Algeria, Bahrain, Egypt, Iraq, Jordan, Kuwait, Lebanon, Libya, Morocco, Oman, Palestine, Qatar, Saudi Arabia, Tunisia, United Arab Emirates, United States, Canada, Nigeria, Kenya, South Africa, Ghana, Ethiopia

Field

Sciences Research

Role type

Specialist

Posted

8/11/2026

Places left

47

We wrote this page from the public micro1 listing. It may be out of date, so read the full posting before you apply.