Use your surveillance investigation experience to build realistic cases that assess whether AI can interpret trade records and employee communications, identify relevant evidence, and reach a defensible conclusion. You’ll create fictional-firm scenarios that test the distinction between potential misconduct and legitimate activity, including cases with incomplete records or competing explanations.
Use your underwriting or claims experience to build insurance cases that test whether AI follows an insurer’s rules and reaches sound decisions. You’ll create fictional files, document the evidence and applicable policy wording, then prepare outputs such as issued policies, coverage letters, or settlement recommendations. The freelance contract runs for an initial three months at about 20 hours per week.
Use your experience with EU regulatory returns to create practical cases that test AI on reporting decisions. You’ll build scenarios for fictional banks, insurers, or other firms, then work through large datasets with spreadsheets and Python to check calculations and confirm the correct template values under the rules in force on a given reporting date.
Apply hands-on industrial process plant experience to assess a simulated facility, its synthetic sensor data, and proposed fault and maintenance scenarios. You’ll judge whether equipment behavior and operating conditions match real plant operations, examine evidence behind diagnoses, and give specific feedback on maintenance decisions, costs, and downtime tradeoffs.
Support a short-term engineering diagram project for an AI research organization by reading and drawing P&IDs according to written conventions. The contract is expected to last 1–2 weeks, with about 20 hours of work in the initial project. Coordinate individual assignments and hours with the project team.
Create and assess benchmarks that test AI systems against real creative and technical workflows in music, audio, speech, linguistics, visual media, or design. The work includes Linux-based tasks and domain-focused review of AI results. The project pays $1,350 per task and expects a commitment of 20 hours per week for 1–6 months.
Use scientific or clinical expertise to assess how AI systems handle domain-specific work. You will create Linux-terminal benchmarks, set standards for verifiable results, and review the accuracy and reasoning behind proposed solutions. The project expects 20 hours per week for 1–6 months and pays $1,350 per task.
Create realistic finance and operations challenges that help benchmark AI systems. You’ll build scenarios from professional workflows, set clear criteria for correct results, and assess proposed answers for accuracy and sound reasoning. The project expects 20 hours of work per week for 1–6 months and pays $1,350 per task.
Use your systems or hardware engineering experience to create difficult, clearly specified challenges for AI agents to solve in a Linux terminal. The project is task-based and expected to last a few weeks. Tasks average six to seven hours, with early submissions potentially taking longer. Earn up to $1,350 per task, with higher rates possible based on task quality and submission volume. Onboarding requires two courses, a graded assessment of about 90 minutes, and a live webinar.
Create difficult benchmark tasks from your professional experience in coding, machine learning, or cybersecurity. Each task must run in a Linux terminal and challenge advanced AI agents: three leading models get three attempts each, and they should fail most attempts. The project should run for the next few weeks. Onboarding requires two courses, a graded assessment of about 90 minutes, and a live webinar before tasking. Pay reaches $1,350 per task, with higher rates possible based on task quality and submission volume; tasks average six to seven hours, and your first tasks may take longer.
France-based contributors fluent or near-native in French will photograph everyday objects with a smartphone, add item details, and provide product URLs when requested. The project targets about 100–400 approved assets each week, with a schedule agreed with the project team and no fixed hours. Payment is $1.50 per approved asset; the stated estimate is about $23 USD per hour, though actual earnings depend on approved submissions. The current deadline is October 30, 2026, and may change with delivery progress.
Create evaluation tasks that ask a model to design and refine passive on-chip RF components against electrical and area targets. For each task, provide the specification, a reference layout, and a check that runs an EM simulation and compares its results with the target. The work starts with five pilot tasks, followed by a review before the project scales. Applicants must be based in Europe, Canada, or the US.
Record short French audio clips, perform brief scripts with varied emotions and speaking styles, and review recordings by noting details in text. The work supports AI systems that interpret speech, tone, and expression. Core project work pays up to $25 USD per hour; rates may vary, and some onboarding, assessment, or non-core work may pay differently. Missions may offer extra incentives.
Bring 3–5 years of QA or quality engineering experience to web and mobile testing for a Web3 product. You’ll combine manual and automated checks, maintain regression coverage, and test wallet, smart contract, and exchange integration flows. The role calls for working-level scripting skills and clear bug reports, with collaboration across US, Europe, and India time zones.
Bring hands-on software engineering judgment to the evaluation of large language models. You’ll create coding benchmarks, assess model responses, and investigate security-related tasks and reasoning failures. The contract starts at six months, with potential for a longer engagement. Paris or London candidates are preferred; strong candidates elsewhere in Europe may work remotely.
Location and eligibility
On-site · Eligible in 10 countries
Engagement type
Contract
Compensation
$75–$90/hr
software engineering
python
data annotation
Machine Learning & AIView job
Showing 15 active roles of 15. Dates reflect what the source published, not when AIWorkJobs last synced.
Source: Braintrust. AIWorkJobs does not verify every claim made by a source. Report a listing that looks wrong through the contact page.