وظائف أخصائي توحد في البحرين

<p><h4>Description</h4> <p>Please submit your CV in English and indicate your level of English proficiency.</p> <p>Mindrift connects specialists with project-based AI opportunities for leading tech companies, focused on testing, evaluating, and improving AI systems. Participation is project-based, not permanent employment.</p> <h4>What this opportunity involves</h4> <p>While each project involves unique tasks, contributors may:</p> <ul> <li>Evaluate AI-generated auto insurance claims decisions for accuracy, coverage correctness, and regulatory compliance;</li> <li>Design FNOL scenarios with deliberate contradictions, decoy files, and outdated documents to test agent robustness;</li> <li>Write and grade fraud-flagging scenarios using structured reason codes (late reporting, recently purchased policy, inconsistent damage) for SIU referral;</li> <li>Build subrogation test cases applying state-specific negligence rules (comparative vs. contributory) and assess likelihood of recovery;</li> <li>Document test cases clearly with correct answers, policy citations, and payout calculations.</li> </ul> <h4>What we look for</h4> <p>This opportunity is a good fit for professionals with a background in insurance claims, legal services, or broader financial services who are open to part-time, non-permanent projects. Ideally, contributors will have:</p> <ul> <li>Degree in Insurance, Risk Management, Business Administration, Finance, Law, or any related field;</li> <li>3+ years of insurance, claims, legal, or financial services experience;</li> <li>Current or recent experience in claims and adjusting or adjacent roles;</li> <li>Familiarity with auto insurance coverage decisions, state-specific negligence rules, and adjuster authority-limit culture;</li> <li>AIC, CPCU, CIFI, or SCLA credential is a strong positive signal, though not required if hands-on experience is solid;</li> <li>Strong written English (C1+).</li> </ul> <h4>How it works</h4> <p>Apply? Pass qualification(s)? Join a project? Complete tasks? Get paid.</p> <h4>Project time expectations</h4> <p>For this project, tasks are estimated to require around 10–20 hours per week during active phases, based on project requirements. This is an estimate, not a guaranteed workload, and applies only while the project is active.</p> <h4>Compensation</h4> <p>On this project, contributors can earn up to $50 per hour equivalent, depending on their level and pace of contribution.</p> <p>Compensation varies across projects depending on scope, complexity, and required expertise. Please note that other projects on the platform may offer different earning levels based on their requirements.</p></p><p></p>

<p><h4>Description</h4> <p>Please submit your CV in English and indicate your level of English proficiency.</p> <p>Mindrift connects specialists with project-based AI opportunities for leading tech companies, focused on testing, evaluating, and improving AI systems. Participation is project-based, not permanent employment.</p> <h4>What this opportunity involves</h4> <p>While each project involves unique tasks, contributors may:</p> <ul> <li>Design and evaluate Quarterly Business Review scenarios — performance vs. benchmark, allocation drift against IPS tolerance bands, and material fund document changes;</li> <li>Create rebalancing and tax-loss harvesting test cases involving cost basis, short-term vs. long-term capital gains, and wash-sale rule application;</li> <li>Develop suitability check cases probing risk tolerance, time horizon, prohibited holdings, and concentration limits against a client's Investment Policy Statement;</li> <li>Construct fee-stack analysis cases: portfolio-weighted expense ratios, advisory fees, and all-in cost comparisons against category benchmarks;</li> <li>Document test cases clearly with verified calculations, policy citations, and correct answers.</li> </ul> <h4>What we look for</h4> <p>This opportunity is a good fit for professionals with a background in investment management, financial advisory, or broader financial services who are open to part-time, non-permanent projects. Ideally, contributors will have:</p> <ul> <li>Degree in finance, economics, business administration, accounting, mathematics, statistics, or any related field;</li> <li>2+ years of finance, investment, accounting, or financial services experience;</li> <li>Current or recent experience in advisory and management or adjacent roles;</li> <li>Practical knowledge of portfolio performance analysis, tax-lot accounting, wash-sale rules, and Investment Policy Statement literacy; a CFA, CFP, CPWA, CIMA, or CIPM credential is a strong positive signal;</li> <li>Strong written English (C1+).</li> </ul> <h4>How it works</h4> <p>Apply ? Pass qualification(s) ? Join a project ? Complete tasks ? Get paid</p> <h4>Project time expectations</h4> <p>For this project, tasks are estimated to require around 10–20 hours per week during active phases, based on project requirements. This is an estimate, not a guaranteed workload, and applies only while the project is active.</p> <h4>Compensation</h4> <p>On this project, contributors can earn up to $50 per hour equivalent, depending on their level and pace of contribution.</p> <p>Compensation varies across projects depending on scope, complexity, and required expertise. Please note that other projects on the platform may offer different earning levels based on their requirements.</p></p><p></p>

<p><h4>Description</h4> <p>Please submit your CV in English and indicate your level of English proficiency.</p> <p>Mindrift connects specialists with project-based AI opportunities for leading tech companies, focused on testing, evaluating, and improving AI systems. Participation is project-based, not permanent employment.</p> <h4>What this opportunity involves</h4> <p>While each project involves unique tasks, contributors may:</p> <ul> <li>Evaluate AI-generated auto insurance claims decisions for accuracy, coverage correctness, and regulatory compliance;</li> <li>Design FNOL scenarios with deliberate contradictions, decoy files, and outdated documents to test agent robustness;</li> <li>Write and grade fraud-flagging scenarios using structured reason codes (late reporting, recently purchased policy, inconsistent damage) for SIU referral;</li> <li>Build subrogation test cases applying state-specific negligence rules (comparative vs. contributory) and assess likelihood of recovery;</li> <li>Document test cases clearly with correct answers, policy citations, and payout calculations.</li> </ul> <h4>What we look for</h4> <p>This opportunity is a good fit for professionals with a background in insurance claims, legal services, or broader financial services who are open to part-time, non-permanent projects. Ideally, contributors will have:</p> <ul> <li>Degree in Insurance, Risk Management, Business Administration, Finance, Law, or any related field;</li> <li>3+ years of insurance, claims, legal, or financial services experience;</li> <li>Current or recent experience in claims and adjusting or adjacent roles;</li> <li>Familiarity with auto insurance coverage decisions, state-specific negligence rules, and adjuster authority-limit culture;</li> <li>AIC, CPCU, CIFI, or SCLA credential is a strong positive signal, though not required if hands-on experience is solid;</li> <li>Strong written English (C1+).</li> </ul> <h4>How it works</h4> <p>Apply > Pass qualification(s) > Join a project > Complete tasks > Get paid</p> <h4>Project time expectations</h4> <p>For this project, tasks are estimated to require around 10–20 hours per week during active phases, based on project requirements. This is an estimate, not a guaranteed workload, and applies only while the project is active.</p> <h4>Compensation</h4> <p>On this project, contributors can earn up to $50 per hour equivalent, depending on their level and pace of contribution.</p> <p>Compensation varies across projects depending on scope, complexity, and required expertise. Please note that other projects on the platform may offer different earning levels based on their requirements.</p></p><p></p>

<p><h4>Description</h4> <p>Please submit your CV in English and indicate your level of English proficiency.</p> <p>Mindrift connects specialists with project-based AI opportunities for leading tech companies, focused on testing, evaluating, and improving AI systems. Participation is project-based, not permanent employment.</p> <h4>What this opportunity involves</h4> <p>While each project involves unique tasks, contributors may:</p> <ul> <li>Evaluate AI-generated mortgage underwriting decisions across loan types (conforming, jumbo, FHA, VA, cash-out refinance);</li> <li>Create test cases involving qualifying income calculations for self-employed borrowers using Schedule C, K-1, and Schedule E;</li> <li>Draft and audit conditional approval letters and adverse action notices for Reg B / ECOA compliance;</li> <li>Distinguish Fannie Mae / Freddie Mac selling guide rules from lender overlays in jumbo and non-conforming loan scenarios;</li> <li>Document problem statements clearly and provide verified correct answers.</li> </ul> <h4>What we look for</h4> <p>This opportunity is a good fit for mortgage underwriters and loan origination professionals open to part-time, non-permanent projects. Ideally, contributors will have:</p> <ul> <li>Degree in Finance, Economics, Business Administration, Accounting, Real Estate, or any related field;</li> <li>3+ years of mortgage, lending, real estate, or financial services experience;</li> <li>Current or recent experience in underwriting & origination or adjacent roles;</li> <li>Ability to calculate qualifying income for self-employed borrowers from IRS Forms 1040, Schedule C, Schedule E, and K-1;</li> <li>Deep familiarity with Fannie Mae / Freddie Mac selling guides and knowledge of Reg B, Reg Z, and TRID disclosure timing rules;</li> <li>Strong written English (C1+).</li> </ul> <h4>How it works</h4> <p>Apply? Pass qualification(s)? Join a project? Complete tasks? Get paid.</p> <h4>Project time expectations</h4> <p>For this project, tasks are estimated to require around 10–20 hours per week during active phases, based on project requirements. This is an estimate, not a guaranteed workload, and applies only while the project is active.</p> <h4>Compensation</h4> <p>On this project, contributors can earn up to $50 per hour equivalent, depending on their level and pace of contribution.</p> <p>Compensation varies across projects depending on scope, complexity, and required expertise. Please note that other projects on the platform may offer different earning levels based on their requirements.</p></p><p></p>

<p><h4>Description</h4> <p>Please submit your CV in English and indicate your level of English proficiency.</p> <p>Mindrift connects specialists with project-based AI opportunities for leading tech companies, focused on testing, evaluating, and improving AI systems. Participation is project-based, not permanent employment.</p> <h4>What this opportunity involves</h4> <p>While each project involves unique tasks, contributors may:</p> <ul> <li>Design and evaluate quarterly business review scenarios — performance vs. benchmark, allocation drift against IPS tolerance bands, and material fund document changes</li> <li>Create rebalancing and tax-loss harvesting test cases involving cost basis, short-term vs. long-term capital gains, and wash-sale rule application</li> <li>Develop suitability check cases probing risk tolerance, time horizon, prohibited holdings, and concentration limits against a client's investment policy statement</li> <li>Construct fee-stack analysis cases: portfolio-weighted expense ratios, advisory fees, and all-in cost comparisons against category benchmarks</li> <li>Document test cases clearly with verified calculations, policy citations, and correct answers</li> </ul> <h4>What we look for</h4> <p>This opportunity is a good fit for professionals with a background in investment management, financial advisory, or broader financial services who are open to part-time, non-permanent projects. Ideally, contributors will have:</p> <ul> <li>Degree in finance, economics, business administration, accounting, mathematics, statistics, or any related field</li> <li>2+ years of finance, investment, accounting, or financial services experience</li> <li>Current or recent experience in advisory and management or adjacent roles</li> <li>Practical knowledge of portfolio performance analysis, tax-lot accounting, wash-sale rules, and investment policy statement literacy; a CFA, CFP, CPWA, CIMA, or CIPM credential is a strong positive signal</li> <li>Strong written English (C1+)</li> </ul> <h4>How it works</h4> <p>Apply ? Pass qualification(s) ? Join a project ? Complete tasks ? Get paid</p> <h4>Project time expectations</h4> <p>For this project, tasks are estimated to require around 10–20 hours per week during active phases, based on project requirements. This is an estimate, not a guaranteed workload, and applies only while the project is active.</p> <h4>Compensation</h4> <p>On this project, contributors can earn up to $50 per hour equivalent, depending on their level and pace of contribution.</p> <p>Compensation varies across projects depending on scope, complexity, and required expertise. Please note that other projects on the platform may offer different earning levels based on their requirements.</p></p><p></p>

<p><h4>Please submit your CV in English and indicate your level of English proficiency.<\/h4>\n<p>Mindrift connects specialists with project-based AI opportunities for leading tech companies, focused on testing, evaluating, and improving AI systems. Participation is project-based, not permanent employment.<\/p>\n<h4>What this opportunity involves:<\/h4>\n<p>We're building a dataset to evaluate AI coding agents - how well a model handles real-world developer tasks.<\/p>\n<p>You'll create challenging tasks and evaluation criteria within realistic simulated environments:<\/p>\n<li>Build realistic developer environments - a virtual company with codebase, infrastructure, and context (tickets, docs, conversations) that forms a believable development history<\/li>\n<li>Design tasks from intermediate states of these environments - craft the prompt, define what \"solved\" means, and ensure the task is solvable by an AI agent<\/li>\n<li>Write tests that verify agent solutions - accept all valid approaches and reject incorrect ones, neither too strict nor too lenient<\/li>\n<li>Iterate on tasks and tests based on QA feedback - review agent solutions, analyze failures, and refine until the evaluation is fair and robust<\/li>\n<h4>What this is not:<\/h4>\n<li>Not data labeling<\/li>\n<li>Not prompt engineering<\/li>\n<li>Not writing code from scratch - the agent writes most of the code; you guide and evaluate<\/li>\n<h4>What we look for:<\/h4>\n<li>8+ years in software development<\/li>\n<li>Core stack: Python (FastAPI), JavaScript\/TypeScript (React), Docker, Postgres, Kafka, Redis<\/li>\n<li>Experience writing tests (functional, integration)<\/li>\n<li>English proficiency - B2+<\/li>\n<h4>Why this is hard:<\/h4>\n<p>Frontier models are already good at coding. Creating a task that genuinely challenges the best models is non-trivial. You need to deeply understand where models fail and what scenarios reveal the difference between a good and a bad solution. Tasks have many valid solutions - writing tests that accept all correct solutions and reject incorrect ones is harder than it sounds.<\/p>\n<h4>How it works<\/h4>\n<p>Apply? Pass qualification(s)? Join a project? Complete tasks? Get paid.<\/p>\n<h4>Effort estimate<\/h4>\n<p>Tasks for this project are estimated to take 30 hours to complete, depending on complexity. This is an estimate and not a schedule requirement; you choose when and how to work. Tasks must be submitted by the deadline and meet the listed acceptance criteria to be accepted.<\/p>\n<h4>Compensation<\/h4>\n<p>Compensation details are provided upon engagement. Tasks are estimated at approximately 30 hours each; you set your own schedule.<\/p><\/p><p><\/p>

<h2 class="h5">Job description</h2> <div class="t-break" data-jb-field="description"> <span>Please submit your CV in English and indicate your level of English proficiency.<br> Mindrift connects specialists with project-based AI opportunities for leading tech companies, focused on testing, evaluating, and improving AI systems.<br> Participation is project-based, not permanent employment.<br> What this opportunity involves While each project involves unique tasks, contributors may: Evaluate AI-generated mortgage underwriting decisions across loan types (conforming, jumbo, FHA, VA, cash-out refinance); Create test cases involving qualifying income calculations for self-employed borrowers using Schedule C, K-1, and Schedule E; Draft and audit conditional approval letters and adverse action notices for Reg B / ECOA compliance; Distinguish Fannie Mae / Freddie Mac selling guide rules from lender overlays in jumbo and non-conforming loan scenarios; Document problem statements clearly and provide verified correct answers.<br> What we look for This opportunity is a good fit for mortgage underwriters and loan origination professionals open to part-time, non-permanent projects.<br> Ideally, contributors will have: Degree in Finance, Economics, Business Administration, Accounting, Real Estate, or any related field; 3+ years of of mortgage, lending, real estate, or financial services experience; Current or recent experience in underwriting &amp; origination or adjacent roles; Ability to calculate qualifying income for self-employed borrowers from IRS Forms 1040, Schedule C, Schedule E, and K-1; Deep familiarity with Fannie Mae / Freddie Mac selling guides and knowledge of Reg B, Reg Z, and TRID disclosure timing rules; Strong written English (C1+).<br> How it works Apply → Pass qualification(s) → Join a project → Complete tasks → Get paid Project time expectations For this project, tasks are estimated to require around 10–20 hours per week during active phases, based on project requirements.<br> This is an estimate, not a guaranteed workload, and applies only while the project is active.<br> Compensation On this project, contributors can earn up to $50 per hour equivalent , depending on their level and pace of contribution.<br> Compensation varies across projects depending on scope, complexity, and required expertise.<br> Please note that other projects on the platform may offer different earning levels based on their requirements.<br></span> </div>

<h2 class="h5">Job description</h2> <div class="t-break" data-jb-field="description"> <span>Please submit your CV in English and indicate your level of English proficiency.<br> Mindrift connects specialists with project-based AI opportunities for leading tech companies, focused on testing, evaluating, and improving AI systems.<br> Participation is project-based, not permanent employment.<br> What this opportunity involves While each project involves unique tasks, contributors may: Evaluate AI-generated auto insurance claims decisions for accuracy, coverage correctness, and regulatory compliance; Design FNOL scenarios with deliberate contradictions, decoy files, and outdated documents to test agent robustness; Write and grade fraud-flagging scenarios using structured reason codes (late reporting, recently purchased policy, inconsistent damage) for SIU referral; Build subrogation test cases applying state-specific negligence rules (comparative vs.<br> contributory) and assess likelihood of recovery; Document test cases clearly with correct answers, policy citations, and payout calculations.<br> What we look for This opportunity is a good fit for professionals with a background in insurance claims, legal services, or broader financial services who are open to part-time, non-permanent projects.<br> Ideally, contributors will have: Degree in Insurance, Risk Management, Business Administration, Finance, Law, or any related field; 3+ years of insurance, claims, legal, or financial services experience; Current or recent experience in claims &amp; adjusting or adjacent roles; Familiarity with auto insurance coverage decisions, state-specific negligence rules, and adjuster authority-limit culture; AIC, CPCU, CIFI, or SCLA credential is a strong positive signal, though not required if hands-on experience is solid; Strong written English (C1+).<br> How it works Apply → Pass qualification(s) → Join a project → Complete tasks → Get paid Project time expectations For this project, tasks are estimated to require around 10–20 hours per week during active phases, based on project requirements.<br> This is an estimate, not a guaranteed workload, and applies only while the project is active.<br> Compensation On this project, contributors can earn up to $50 per hour equivalent , depending on their level and pace of contribution.<br> Compensation varies across projects depending on scope, complexity, and required expertise.<br> Please note that other projects on the platform may offer different earning levels based on their requirements.<br></span> </div>

<h2 class="h5">Job description</h2> <div class="t-break" data-jb-field="description"> <span>Please submit your CV in English and indicate your level of English proficiency.<br> Mindrift connects specialists with project-based AI opportunities for leading tech companies, focused on testing, evaluating, and improving AI systems.<br> Participation is project-based, not permanent employment.<br> What this opportunity involves While each project involves unique tasks, contributors may: Design and evaluate Quarterly Business Review scenarios — performance vs.<br> benchmark, allocation drift against IPS tolerance bands, and material fund document changes; Create rebalancing and tax-loss harvesting test cases involving cost basis, short-term vs.<br> long-term capital gains, and wash-sale rule application; Develop suitability check cases probing risk tolerance, time horizon, prohibited holdings, and concentration limits against a client's Investment Policy Statement; Construct fee-stack analysis cases: portfolio-weighted expense ratios, advisory fees, and all-in cost comparisons against category benchmarks; Document test cases clearly with verified calculations, policy citations, and correct answers.<br> What we look for This opportunity is a good fit for professionals with a background in investment management, financial advisory, or broader financial services who are open to part-time, non-permanent projects.<br> Ideally, contributors will have: Degree in Finance, Economics, Business Administration, Accounting, Mathematics, Statistics, or any related field; 2+ years of finance, investment, accounting, or financial services experience; Current or recent experience in advisory &amp; management or adjacent roles; Practical knowledge of portfolio performance analysis, tax-lot accounting, wash-sale rules, and Investment Policy Statement literacy; a CFA, CFP, CPWA, CIMA, or CIPM credential is a strong positive signal; Strong written English (C1+).<br> How it works Apply → Pass qualification(s) → Join a project → Complete tasks → Get paid Project time expectations For this project, tasks are estimated to require around 10–20 hours per week during active phases, based on project requirements.<br> This is an estimate, not a guaranteed workload, and applies only while the project is active.<br> Compensation On this project, contributors can earn up to $50 per hour equivalent , depending on their level and pace of contribution.<br> Compensation varies across projects depending on scope, complexity, and required expertise.<br> Please note that other projects on the platform may offer different earning levels based on their requirements.<br></span> </div>

<p><h4>Description</h4> <p>Please submit your CV in English and indicate your level of English proficiency.</p> <p>Mindrift connects specialists with project-based AI opportunities for leading tech companies, focused on testing, evaluating, and improving AI systems. Participation is project-based, not permanent employment.</p> <h4>What this opportunity involves</h4> <p>While each project involves unique tasks, contributors may:</p> <ul> <li>Design and evaluate rule application scenarios — reading a compliance artifact (advertisement, disclosure, policy) against a specific regulation and citing the exact subsection that applies or is violated;</li> <li>Write sanctions alert disposition cases with structured false-positive reasoning documentation and escalation scenarios;</li> <li>Build compliance testing scenarios: sample selection, test design, findings documentation, and examiner-ready conclusion memos;</li> <li>Develop policy gap identification cases where an internal policy is silent on a regulatory requirement, contradicts the regulation, or has not been updated to reflect a recent change;</li> <li>Document all test cases with verified regulatory citations to the specific subsection — never approximated, never fabricated.</li> </ul> <h4>What we look for</h4> <p>This opportunity is a good fit for professionals with a background in compliance, regulatory affairs, legal services, or broader financial services who are open to part-time, non-permanent projects. Ideally, contributors will have:</p> <ul> <li>Degree in law, finance, economics, business administration, accounting, or any related field;</li> <li>2+ years of compliance, audit, legal, risk management, or financial services experience;</li> <li>Current or recent experience in compliance and regulatory roles, financial crimes, or adjacent roles;</li> <li>Citation discipline as a core habit — able to identify the exact regulatory subsection that applies and flag uncertainty rather than approximate;</li> <li>CRCM, CAMS, or equivalent local credential is a strong positive signal; former regulatory examiner or inspector status is equally strong;</li> <li>Strong written English (C1+).</li> </ul> <h4>How it works</h4> <p>Apply? Pass qualification(s)? Join a project? Complete tasks? Get paid.</p> <h4>Project time expectations</h4> <p>For this project, tasks are estimated to require around 10–20 hours per week during active phases, based on project requirements. This is an estimate, not a guaranteed workload, and applies only while the project is active.</p> <h4>Compensation</h4> <p>On this project, contributors can earn up to $50 per hour equivalent, depending on their level and pace of contribution.</p> <p>Compensation varies across projects depending on scope, complexity, and required expertise. Please note that other projects on the platform may offer different earning levels based on their requirements.</p></p><p></p>

<p><h4>Description</h4> <p>Please submit your CV in English and indicate your level of English proficiency.</p> <p>Mindrift connects specialists with project-based AI opportunities for leading tech companies, focused on testing, evaluating, and improving AI systems. Participation is project-based, not permanent employment.</p> <h4>What this opportunity involves</h4> <p>While each project involves unique tasks, contributors may:</p> <ul> <li>Design and evaluate rule application scenarios — reading a compliance artifact (advertisement, disclosure, policy) against a specific regulation and citing the exact subsection that applies or is violated;</li> <li>Write sanctions alert disposition cases with structured false-positive reasoning documentation and escalation scenarios;</li> <li>Build compliance testing scenarios: sample selection, test design, findings documentation, and examiner-ready conclusion memos;</li> <li>Develop policy gap identification cases where an internal policy is silent on a regulatory requirement, contradicts the regulation, or has not been updated to reflect a recent change;</li> <li>Document all test cases with verified regulatory citations to the specific subsection — never approximated, never fabricated.</li> </ul> <h4>What we look for</h4> <p>This opportunity is a good fit for professionals with a background in compliance, regulatory affairs, legal services, or broader financial services who are open to part-time, non-permanent projects. Ideally, contributors will have:</p> <ul> <li>Degree in Law, Finance, Economics, Business Administration, Accounting, or any related field;</li> <li>2+ years of compliance, audit, legal, risk management, or financial services experience;</li> <li>Current or recent experience in compliance and regulatory roles, financial crimes, or adjacent roles;</li> <li>Citation discipline as a core habit — able to identify the exact regulatory subsection that applies and flag uncertainty rather than approximate;</li> <li>CRCM, CAMS, or equivalent local credential is a strong positive signal; former regulatory examiner or inspector status is equally strong;</li> <li>Strong written English (C1+).</li> </ul> <h4>How it works</h4> <p>Apply ? Pass qualification(s) ? Join a project ? Complete tasks ? Get paid</p> <h4>Project time expectations</h4> <p>For this project, tasks are estimated to require around 10–20 hours per week during active phases, based on project requirements. This is an estimate, not a guaranteed workload, and applies only while the project is active.</p> <h4>Compensation</h4> <p>On this project, contributors can earn up to $50 per hour equivalent, depending on their level and pace of contribution.</p> <p>Compensation varies across projects depending on scope, complexity, and required expertise. Please note that other projects on the platform may offer different earning levels based on their requirements.</p></p><p></p>

<p><h4>Please submit your CV in English and indicate your level of English proficiency.<\/h4>\n<p>Mindrift connects specialists with project-based AI opportunities for leading tech companies, focused on testing, evaluating, and improving AI systems. Participation is project-based, not permanent employment.<\/p>\n<h4>What this opportunity involves:<\/h4>\n<p>We're building a dataset to evaluate AI coding agents - how well a model handles real-world developer tasks.<\/p>\n<p>You'll create challenging tasks and evaluation criteria within realistic simulated environments:<\/p>\n<ul>\n <li>Build realistic developer environments - a virtual company with codebase, infrastructure, and context (tickets, docs, conversations) that forms a believable development history<\/li>\n <li>Design tasks from intermediate states of these environments - craft the prompt, define what \"solved\" means, and ensure the task is solvable by an AI agent<\/li>\n <li>Write tests that verify agent solutions - accept all valid approaches and reject incorrect ones, neither too strict nor too lenient<\/li>\n <li>Iterate on tasks and tests based on QA feedback - review agent solutions, analyze failures, and refine until the evaluation is fair and robust<\/li>\n<\/ul>\n<h4>What this is NOT:<\/h4>\n<ul>\n <li>Not data labeling<\/li>\n <li>Not prompt engineering<\/li>\n <li>Not writing code from scratch - the agent writes most of the code; you guide and evaluate<\/li>\n<\/ul>\n<h4>What we look for:<\/h4>\n<ul>\n <li>8+ years in software development<\/li>\n <li>Core stack: Python (FastAPI), JavaScript\/TypeScript (React), Docker, Postgres, Kafka, Redis<\/li>\n <li>Experience writing tests (functional, integration)<\/li>\n <li>English proficiency - B2+<\/li>\n<\/ul>\n<h4>Why this is hard:<\/h4>\n<p>Frontier models are already good at coding. Creating a task that genuinely challenges the best models is non-trivial. You need to deeply understand where models fail and what scenarios reveal the difference between a good and a bad solution. Tasks have many valid solutions - writing tests that accept all correct solutions and reject incorrect ones is harder than it sounds.<\/p>\n<h4>How it works<\/h4>\n<p>Apply? Pass qualification(s)? Join a project? Complete tasks? Get paid.<\/p>\n<h4>Effort estimate<\/h4>\n<p>Tasks for this project are estimated to take 30 hours to complete, depending on complexity. This is an estimate and not a schedule requirement; you choose when and how to work. Tasks must be submitted by the deadline and meet the listed acceptance criteria to be accepted.<\/p>\n<h4>Compensation:<\/h4>\n<p>Up to $200\/hr equivalent, depending on level and pace. Tasks are estimated at approximately 30 hours each; you set your own schedule.<\/p><\/p><p><\/p>

<h2 class="h5">Job description</h2> <div class="t-break" data-jb-field="description"> <span>Please submit your CV in English and indicate your level of English proficiency.<br> Mindrift connects specialists with project-based AI opportunities for leading tech companies, focused on testing, evaluating, and improving AI systems.<br> Participation is project-based, not permanent employment.<br> What this opportunity involves While each project involves unique tasks, contributors may: Design and evaluate rule application scenarios — reading a compliance artifact (advertisement, disclosure, policy) against a specific regulation and citing the exact subsection that applies or is violated; Write sanctions alert disposition cases with structured false-positive reasoning documentation and escalation scenarios; Build compliance testing scenarios: sample selection, test design, findings documentation, and examiner-ready conclusion memos; Develop policy gap identification cases where an internal policy is silent on a regulatory requirement, contradicts the regulation, or has not been updated to reflect a recent change; Document all test cases with verified regulatory citations to the specific subsection — never approximated, never fabricated.<br> What we look for This opportunity is a good fit for professionals with a background in compliance, regulatory affairs, legal services, or broader financial services who are open to part-time, non-permanent projects.<br> Ideally, contributors will have: Degree in Law, Finance, Economics, Business Administration, Accounting, or any related field; 2+ years of compliance, audit, legal, risk management, or financial services experience; Current or recent experience in compliance &amp; regulatory roles, financial crimes, or adjacent roles; Citation discipline as a core habit — able to identify the exact regulatory subsection that applies and flag uncertainty rather than approximate; CRCM, CAMS, or equivalent local credential is a strong positive signal; former regulatory examiner or inspector status is equally strong; Strong written English (C1+).<br> How it works Apply → Pass qualification(s) → Join a project → Complete tasks → Get paid Project time expectations For this project, tasks are estimated to require around 10–20 hours per week during active phases, based on project requirements.<br> This is an estimate, not a guaranteed workload, and applies only while the project is active.<br> Compensation On this project, contributors can earn up to $50 per hour equivalent , depending on their level and pace of contribution.<br> Compensation varies across projects depending on scope, complexity, and required expertise.<br> Please note that other projects on the platform may offer different earning levels based on their requirements.<br></span> </div>

<h2 class="h5">Job description</h2> <div class="t-break" data-jb-field="description"> <span>Please submit your CV in English and indicate your level of English proficiency.<br> Mindrift connects specialists with project-based AI opportunities for leading tech companies, focused on testing, evaluating, and improving AI systems.<br> Participation is project-based, not permanent employment.<br> What this opportunity involves We're building competitive-programming problems and their automatic graders (test cases + checkers) used to train advanced AI coding models.<br> You'll create problems, write both correct and deliberately-wrong solutions, and design the tough test cases that catch subtly-broken code.<br> Your work is what makes the training signal trustworthy.<br> To apply, share: your competitive-programming handle(s) + current/peak rating, any problem-setting experience, and IOI/ICPC/olympiad history.<br> What you'll do Write original algorithmic problems — clear statement, sound constraints, an intended solution; Write correct solutions in C++ and Python; Write realistic wrong solutions (common mistakes) to test against; Build and harden test cases: generators, edge cases, stress and "hacking" tests; Write checkers/interactors (testlib) for problems with more than one valid answer; Confirm the quality bar: correct solutions pass, wrong ones fail, within time limits.<br> What we look for Strong competitive programming background, shown by a public profile — a Codeforces rating (see levels below), OR an IOI / ICPC / national-olympiad record; Fluent contest C++ (STL, complexity); comfortable writing Python; Can explain why a solution is wrong and build an input that breaks it; Strong written English (C1+).<br> Levels (by Codeforces rating, or equivalent olympiad achievement): Associate — 1700–2100 (Expert / Candidate Master): easier problems; Expert — 2200–2600 (Master / Grandmaster): harder problems Senior / Reviewer — 2600+ (Int'l Grandmaster) or IOI/ICPC medalist: hardest problems + quality review.<br> Nice to have IOI / ICPC alumni and top competitive programmers especially encouraged to apply; Experience setting or testing problems for real contests (Codeforces rounds, ICPC, national olympiads, online judges); Testlib experience (checkers, validators, generators); Experience creating data for LLM code benchmarks or similar RL / evaluation datasets.<br> How it works Apply → Pass qualification(s) → Join a project → Complete tasks → Get paid Project time expectations For this project, tasks are estimated to require around 10-20 hours per week during active phases, based on project requirements.<br> This is an estimate, not a guaranteed workload, and applies only while the project is active.<br> Tasks must be submitted by the deadline and meet the listed acceptance criteria to be accepted.<br> Compensation Paid per accepted task.<br> Your rate depends on the qualification tier you reach and how efficiently you complete tasks — up to the equivalent of $90/hr .<br> Because payment is per task, a faster pace raises your effective hourly rate.<br></span> </div>