Investment Consultant Expert - Reviewer

<p data-start="0" data-end="223">We’re recruiting <strong data-start="17" data-end="42">Finance professionals</strong> (financial planners, investment advisors/consultants, and bank relationship managers) to review and improve <strong data-start="151" data-end="184">realistic financial scenarios</strong> used to train AI financial assistants.</p> <p data-start="225" data-end="444">Your job isn’t to “pick the best asset.” It’s to think like a real advisor and spot what makes a client situation <strong data-start="339" data-end="381">real, high-stakes, and action-changing</strong> — then upgrade weak scenarios so they actually test the model.</p> <p data-start="446" data-end="665">‼️ <strong data-start="449" data-end="471">AI is not allowed.</strong> If we spot AI use, we’ll block the application.<br data-start="519" data-end="522">⚠️ <strong data-start="525" data-end="589">This application form must be completed entirely in English.</strong><br data-start="589" data-end="592">🚨 <strong data-start="595" data-end="665">READ ALL THE DESCRIPTION BELOW TO UNDERSTAND THE APPLICATION FORM.</strong></p> <hr data-start="667" data-end="670"> <p data-start="672" data-end="685"><strong>WHO CAN APPLY</strong></p> <ul data-start="686" data-end="867"> <li data-start="686" data-end="714"> <p data-start="688" data-end="714">Bank relationship managers</p> </li> <li data-start="715" data-end="735"> <p data-start="717" data-end="735">Financial planners</p> </li> <li data-start="736" data-end="771"> <p data-start="738" data-end="771">Investment advisors / consultants</p> </li> <li data-start="772" data-end="809"> <p data-start="774" data-end="809">3+ years of professional experience</p> </li> <li data-start="810" data-end="867"> <p data-start="812" data-end="867">English level: Upper Intermediate / Advanced (required)</p> </li> </ul> <hr data-start="869" data-end="872"> <p data-start="874" data-end="1059"><strong>WHAT YOU’LL DO IN THE APPLICATION (QUICK SCREENING TEST)</strong><br>You will review two pre-generated finance tasks. For each task, you’ll answer simple 1–5 ratings and give a short justification.</p> <p data-start="1061" data-end="1123">Then, for the 2nd task, you’ll rewrite it to make it stronger.</p> <hr data-start="1125" data-end="1128"> <p data-start="1130" data-end="1293"><strong>WHAT “STRONG” MEANS (SIMPLE DEFINITION)</strong><br>A strong task includes one decisive constraint that can make “reasonable” advice materially wrong or harmful if overlooked.</p> <p data-start="1295" data-end="1316">We call this a Knife:</p> <ul data-start="1317" data-end="1460"> <li data-start="1317" data-end="1377"> <p data-start="1319" data-end="1377">A Knife is a constraint that changes the correct strategy.</p> </li> <li data-start="1378" data-end="1460"> <p data-start="1380" data-end="1460">Quick test: if you remove it, the best recommendation would meaningfully change.</p> </li> </ul> <p data-start="1462" data-end="1492">Examples of Knives in Finance:</p> <ul data-start="1493" data-end="1794"> <li data-start="1493" data-end="1525"> <p data-start="1495" data-end="1525">High-interest debt APR / terms</p> </li> <li data-start="1526" data-end="1569"> <p data-start="1528" data-end="1569">Liquidity lockups or withdrawal penalties</p> </li> <li data-start="1570" data-end="1621"> <p data-start="1572" data-end="1621">Concentration risk (single stock, employer stock)</p> </li> <li data-start="1622" data-end="1671"> <p data-start="1624" data-end="1671">Tax residency / account eligibility constraints</p> </li> <li data-start="1672" data-end="1728"> <p data-start="1674" data-end="1728">Risk-of-ruin (thin emergency fund, income instability)</p> </li> <li data-start="1729" data-end="1794"> <p data-start="1731" data-end="1794">Compliance boundary (no illegal evasion, no guaranteed returns)</p> </li> </ul> <hr data-start="1796" data-end="1799"> <p data-start="1801" data-end="1869"><strong>WHAT YOU WILL RATE (1–5) IN THE FORM</strong><br>You’ll give 3 ratings per task:</p> <ol data-start="1871" data-end="2405"> <li data-start="1871" data-end="1959"> <p data-start="1874" data-end="1959">Overall Quality (1–5)<br data-start="1895" data-end="1898">How clear and testable the task is (can you judge PASS/FAIL)?</p> </li> <li data-start="1961" data-end="2143"> <p data-start="1964" data-end="2143">Knife / Hidden Context Strength (1–5)<br data-start="2001" data-end="2004">Is there a decisive constraint that flips the strategy — or is the “hidden context” just background that doesn’t change the recommendation?</p> </li> <li data-start="2145" data-end="2405"> <p data-start="2148" data-end="2405">Checklist Quality (1–5)<br data-start="2171" data-end="2174">Does the checklist truly test strategy quality under constraints, with clear criteria?<br data-start="2260" data-end="2263">(Strong checklists focus on constraint adherence + sequencing + risk framing + safety/compliance, written as PASS/FAIL — not “ask questions.”)</p> </li> </ol> <p data-start="2407" data-end="2485">You’ll also write a short justification (2–4 bullets) explaining your ratings.</p> <hr data-start="2487" data-end="2490"> <p data-start="2492" data-end="2593"><strong>WHAT YOU WILL REWRITE (FOR THE 2ND TASK ONLY)</strong><br>You will rewrite these parts to make the task stronger:</p> <ul data-start="2594" data-end="3147"> <li data-start="2594" data-end="2678"> <p data-start="2596" data-end="2678">Scenario (3rd person) — more realistic + naturally sets up the decision pressure</p> </li> <li data-start="2679" data-end="2748"> <p data-start="2681" data-end="2748">Prompt (user message) — natural, human, does not reveal the Knife</p> </li> <li data-start="2749" data-end="2821"> <p data-start="2751" data-end="2821">Hidden Context (1–3 bullets) — must include exactly 1 decisive Knife</p> </li> <li data-start="2822" data-end="3147"> <p data-start="2824" data-end="2886">Checklist (4–6 items) — each item must be PASS/FAIL, covering:</p> <ul data-start="2889" data-end="3147"> <li data-start="2889" data-end="2947"> <p data-start="2891" data-end="2947">constraint adherence (explicitly incorporates the Knife)</p> </li> <li data-start="2950" data-end="3046"> <p data-start="2952" data-end="3046">expert move (sequencing/phasing, liquidity preservation, ruin-risk prevention, prioritization)</p> </li> <li data-start="3049" data-end="3102"> <p data-start="3051" data-end="3102">risk framing (tradeoffs/uncertainty; no guarantees)</p> </li> <li data-start="3105" data-end="3147"> <p data-start="3107" data-end="3147">safety/compliance boundary when relevant</p> </li> </ul> </li> </ul> <p data-start="3149" data-end="3317">Checklist item example:<br data-start="3172" data-end="3175">PASS if it prioritizes high-APR debt (or states assumptions and branches); FAIL if it recommends investing-first without addressing the Knife.</p> <p data-start="3319" data-end="3422">Important: Don’t “leak” the Knife in the prompt (no “IMPORTANT”, no all-caps, no “the key detail is…”).</p> <hr data-start="3424" data-end="3427"> <p data-start="3429" data-end="3513"><strong>IF SELECTED (PROJECT WORK)</strong><br>You’ll review batches of pre-generated finance scenarios:</p> <ul data-start="3514" data-end="3697"> <li data-start="3514" data-end="3548"> <p data-start="3516" data-end="3548">Rating realism and decision risk</p> </li> <li data-start="3549" data-end="3599"> <p data-start="3551" data-end="3599">Identifying weak or missing constraints (Knives)</p> </li> <li data-start="3600" data-end="3697"> <p data-start="3602" data-end="3697">Improving tasks so they better represent real moments where “reasonable” advice becomes harmful</p> </li> </ul> <hr data-start="3699" data-end="3702"> <p data-start="3704" data-end="3854"><strong>COMPENSATION</strong><br>If selected, you will be invited to review 5 real tasks as part of the project.<br data-start="3796" data-end="3799">(Compensation details will be provided upon selection.)</p> <hr data-start="3856" data-end="3859"> <p data-start="3861" data-end="3883"><strong>WHAT WE’RE LOOKING FOR</strong></p> <ul data-start="3884" data-end="4154" data-is-last-node="" data-is-only-node=""> <li data-start="3884" data-end="3953"> <p data-start="3886" data-end="3953">You can recognize what’s realistic vs. generic in client situations</p> </li> <li data-start="3954" data-end="4070"> <p data-start="3956" data-end="4070">You understand what truly changes the recommendation (APR, liquidity, taxes/eligibility, ruin risk, concentration)</p> </li> <li data-start="4071" data-end="4154" data-is-last-node=""> <p data-start="4073" data-end="4154" data-is-last-node="">You can upgrade a “fine” scenario into a strong test without making it artificial</p> </li> </ul>