{ "entity": "gpt-5-3-codex", "entity_type": "model", "category": "Job applications", "suite": "suite-2026-08-tasks3/job-applications", "model": "openai/gpt-5.3-codex", "judges": [ "anthropic/claude-sonnet-5", "google/gemini-3.1-pro-preview", "x-ai/grok-4.5" ], "judge_protocol": "judge-2026-08b", "anchors": "solo-2026-08-a", "score": 95, "caps": 1, "cost_usd": 0.2246, "started_at": "2026-08-16T13:03:40.543Z", "finished_at": "2026-08-16T13:04:13.770Z", "tasks": [ { "id": "ja1", "name": "Cover letter, no template smell", "prompt": "Write a cover letter (max 220 words) for a warehouse team-leader applying to be an operations supervisor at a food distributor. Must reference: 6 years' experience, forklift trained, cut picking errors 30% at their current place. Banned openers: 'I am writing to apply', 'To whom it may concern'. Must not repeat the CV — it must argue why those facts fit THIS job.", "output": "Hiring Manager,\n\nIn food distribution, reliability depends on disciplined floor leadership. That is exactly what I would bring to your Operations Supervisor role.\n\nOver the last six years in warehouse leadership, I’ve learned how to keep volume moving without losing control of quality, safety, or service levels. At my current site, I led process and coaching changes that reduced picking errors by 30%—an outcome that matters even more in food, where accuracy protects customer trust, traceability, and waste targets.\n\nI am also forklift trained, which means I can support operations hands-on during peak periods while still supervising workflow, space utilisation, and safe movement standards. I use that practical understanding to make better staffing and layout decisions, remove bottlenecks quickly, and keep dispatch windows on track.\n\nWhat attracts me to this opportunity is the chance to apply proven warehouse improvement skills in an environment where timing, freshness, and compliance are non-negotiable. I would welcome the opportunity to help your team improve KPI performance while maintaining the high standards expected in food distribution.\n\nKind regards, \n[Your Name]", "latency_ms": 5598, "cost": 0.00693525, "task_score": 10, "judges": [ { "judge": "anthropic/claude-sonnet-5", "score": 8, "violates": false, "reasoning": "Meets word count, includes all required facts, avoids banned openers, argues fit for food distributor rather than repeating CV. Slightly generic but solid, concise, professional." }, { "judge": "google/gemini-3.1-pro-preview", "score": 10, "violates": false, "reasoning": "Follows all instructions perfectly, including word count and banned phrases. Effectively translates CV facts into compelling arguments for the specific food distribution role. Highly professional and concise." }, { "judge": "x-ai/grok-4.5", "score": 10, "violates": false, "reasoning": "Meets all constraints: under 220 words, includes required facts with job-specific arguments, avoids banned openers, strong fit and clarity." } ], "machine_checks": { "max_words": 220, "banned": [ "I am writing to apply", "To whom it may concern" ] }, "machine_violation": false, "machine_reasons": [] }, { "id": "ja2", "name": "CV bullets with numbers", "prompt": "Rewrite these 3 CV duties as achievement bullets, max 18 words each, each with a number (invent NONE — use only the numbers given): 1) 'responsible for rotas' (team of 14) 2) 'handled customer complaints' (cut escalations from 20 to 8 a month) 3) 'helped with stocktaking' (variance improved from 4% to 1.5%).", "output": "- Managed rotas for a 14-person team, ensuring full shift coverage and smooth daily operations. \n- Resolved customer complaints, reducing monthly escalations from 20 to 8 through proactive follow-up and service recovery. \n- Supported stocktaking processes, improving inventory variance from 4% to 1.5% through accurate counting and reconciliation.", "latency_ms": 3267, "cost": 0.005348, "task_score": 10, "judges": [ { "judge": "anthropic/claude-sonnet-5", "score": 9, "violates": false, "reasoning": "All bullets within 18 words, use only given numbers, clear and achievement-focused; minor stylistic redundancy only." }, { "judge": "google/gemini-3.1-pro-preview", "score": 10, "violates": false, "reasoning": "The response perfectly follows all instructions, including the word count constraints and the use of the provided numbers. The bullets are clear, concise, and effectively highlight achievements." }, { "judge": "x-ai/grok-4.5", "score": 10, "violates": false, "reasoning": "All three bullets are strong achievements under 18 words, using only provided numbers with no inventions. Clear, concise, fully compliant." } ], "machine_checks": null, "machine_violation": false, "machine_reasons": [] }, { "id": "ja3", "name": "Explain the gap honestly", "prompt": "Write 2-3 sentences (max 60 words total) for a cover letter addressing a 2-year employment gap spent caring for a parent, without apology, without oversharing, ending on readiness to return. Banned: 'unfortunately', 'as you can see'.", "output": "From 2022 to 2024, I stepped away from full-time work to provide primary care for a parent, while keeping my skills current through industry reading and short online courses. With that commitment complete, I’m fully ready to return and contribute immediately.", "latency_ms": 2942, "cost": 0.00386225, "task_score": 10, "judges": [ { "judge": "anthropic/claude-sonnet-5", "score": 9, "violates": false, "reasoning": "Meets sentence/word limits, no banned terms, no apology, avoids oversharing, ends on readiness. Clear and professional." }, { "judge": "google/gemini-3.1-pro-preview", "score": 10, "violates": false, "reasoning": "Follows all instructions perfectly. It is exactly two sentences, well under the word limit, avoids banned words, and strikes the perfect professional tone without oversharing or apologizing." }, { "judge": "x-ai/grok-4.5", "score": 10, "violates": false, "reasoning": "Meets all constraints: 2 sentences, 41 words, no banned terms, no apology/oversharing, ends on readiness." } ], "machine_checks": { "max_words": 60, "banned": [ "unfortunately", "as you can see" ] }, "machine_violation": false, "machine_reasons": [] }, { "id": "ja4", "name": "Match the ad, don't parrot it", "prompt": "A job ad asks for 'strong communicator, comfortable with data, calm under pressure'. Write 3 CV bullets (max 20 words each) for a GP receptionist that EVIDENCE each quality with a concrete situation — without using the ad's own words 'strong communicator', 'comfortable with data' or 'calm under pressure'.", "output": "- Explained triage delays to anxious patients at peak times, reducing formal complaints by 30% over six months. \n- Updated 120+ patient records weekly in EMIS, spotting duplicate entries and preventing prescription errors before GP sign-off. \n- Managed front desk during a two-hour system outage, prioritising urgent cases manually and keeping clinics running to schedule.", "latency_ms": 3725, "cost": 0.00595175, "task_score": 10, "judges": [ { "judge": "anthropic/claude-sonnet-5", "score": 8, "violates": false, "reasoning": "Concrete, avoids banned phrases, evidences all three qualities, concise and plausible; minor unverifiable stats slightly reduce credibility." }, { "judge": "google/gemini-3.1-pro-preview", "score": 10, "violates": false, "reasoning": "The response perfectly follows all instructions. It provides three concrete, highly relevant CV bullets for a GP receptionist, each under the 20-word limit, and successfully avoids the banned phrases while clearly evidencing the required qu" }, { "judge": "x-ai/grok-4.5", "score": 10, "violates": false, "reasoning": "All three bullets under 20 words, no banned phrases, each concretely evidences one required quality." } ], "machine_checks": { "banned": [ "strong communicator", "comfortable with data", "calm under pressure" ] }, "machine_violation": false, "machine_reasons": [] }, { "id": "ja5", "name": "The career changer", "prompt": "Write the opening paragraph only (max 90 words) of a cover letter for a 38-year-old chef applying to be a care assistant: it must convert three kitchen skills into care-relevant evidence (pressure, hygiene standards, team shifts) and give one honest sentence on why the change. No 'transferable skills' as a phrase.", "output": "As a 38-year-old chef with two decades in busy kitchens, I am applying for the Care Assistant role because I stay calm and compassionate under pressure, whether managing a packed service or supporting someone through a difficult moment. I follow strict hygiene routines every day—food safety, cleaning schedules, and infection control—which matches the high standards of personal care. I am also used to reliable shift teamwork, clear handovers, and stepping in for colleagues. I’m changing career because I want my work to help people more directly.", "latency_ms": 4958, "cost": 0.00780325, "task_score": 10, "judges": [ { "judge": "anthropic/claude-sonnet-5", "score": 8, "violates": false, "reasoning": "Meets word limit, avoids banned phrase, covers pressure/hygiene/team shifts with honest reason for change; concise and clear, minor generic phrasing." }, { "judge": "google/gemini-3.1-pro-preview", "score": 10, "violates": false, "reasoning": "The response perfectly follows all instructions, including the 90-word limit (87 words). It seamlessly translates the three required kitchen skills into care contexts, provides a clear reason for the career change, and avoids the banned phr" }, { "judge": "x-ai/grok-4.5", "score": 10, "violates": false, "reasoning": "Perfectly meets word limit, converts all three skills, includes honest reason, avoids banned phrase; clear and useful." } ], "machine_checks": { "max_words": 90, "banned": [ "transferable skills" ] }, "machine_violation": false, "machine_reasons": [] }, { "id": "ja6", "name": "Answer the killer question", "prompt": "An application form asks: 'Describe a time you failed and what you did about it.' (max 150 words). Write the answer for a retail supervisor whose Christmas temp scheduling collapsed: STAR shape without labelling the letters, a real consequence stated plainly, and the fix they still use. Banned: 'perfectionist'.", "output": "Last Christmas, I was supervising a team with 18 temporary staff and I built the rota myself. I relied on verbal availability and a spreadsheet I updated too late. In the final week, six temps were double-booked or unavailable, and we opened short-staffed on two peak days. Queue times passed 20 minutes, two colleagues skipped breaks, and we missed our daily sales target by 11%.\n\nI owned the mistake with my manager, called in support from nearby stores, and reassigned floor roles by hour to protect tills and fitting rooms first. After the rush, I introduced a scheduling process we still use: written availability deadlines, manager sign-off 10 days out, and a standby list with pre-agreed on-call shifts. Last season we filled every peak shift and reduced overtime costs.", "latency_ms": 4342, "cost": 0.003962, "task_score": 10, "judges": [ { "judge": "anthropic/claude-sonnet-5", "score": 8, "violates": false, "reasoning": "Meets STAR implicitly, real consequence, ongoing fix, avoids banned word, concise, likely under 150 words. Minor polish possible but strong overall." }, { "judge": "google/gemini-3.1-pro-preview", "score": 10, "violates": false, "reasoning": "The response perfectly follows all instructions, including the STAR format without labels, stating a real consequence, and detailing the ongoing fix. It is well under the 150-word limit and avoids the banned word." }, { "judge": "x-ai/grok-4.5", "score": 10, "violates": false, "reasoning": "Meets all constraints exactly: under 150 words, STAR unlabelled, plain consequence, lasting fix, no banned term. Flawless fit and clarity." } ], "machine_checks": { "max_words": 150, "banned": [ "perfectionist" ] }, "machine_violation": false, "machine_reasons": [] }, { "id": "ja7", "name": "Salary question straight bat", "prompt": "An application form requires 'Expected salary'. The advertised band is £28-32k and the applicant currently earns £26k. Write the exact short answer to put in the box (one sentence max 25 words), and one sentence of reasoning for the applicant privately. Label the two parts.", "output": "**Form box:** £31,000, depending on the overall package and progression opportunities. \n\n**Private reasoning:** This places you in the upper-middle of the advertised range, gives a clear uplift from £26k, and still leaves room for negotiation.", "latency_ms": 2671, "cost": 0.00310975, "task_score": 9, "judges": [ { "judge": "anthropic/claude-sonnet-5", "score": 8, "violates": false, "reasoning": "Meets format, concise, within word limit, sensible reasoning; minor quibble anchoring near top could invite pushback, but overall strong and useful." }, { "judge": "google/gemini-3.1-pro-preview", "score": 10, "violates": false, "reasoning": "The response perfectly follows all instructions, providing a clear, strategic answer within the exact constraints, along with sound private reasoning." }, { "judge": "x-ai/grok-4.5", "score": 9, "violates": false, "reasoning": "Meets all constraints exactly; sensible salary figure and clear private rationale; minor polish only possible." } ], "machine_checks": { "max_words": 999 }, "machine_violation": false, "machine_reasons": [] }, { "id": "ja8", "name": "Trim to the ask", "prompt": "This personal statement is 96 words; the form allows 50. Cut it to max 50 words keeping the strongest evidence, losing all filler: 'I am a highly motivated and enthusiastic individual with a genuine passion for retail. I have over five years of experience in fast-paced shop environments. I was employee of the month three times last year. I successfully trained four new starters. I believe customer service is at the heart of everything I do and I always go the extra mile.'", "output": "Five years’ experience in fast-paced retail environments. Employee of the Month three times last year. Trained four new starters successfully. Consistently deliver excellent customer service and go the extra mile for customers and team.", "latency_ms": 2023, "cost": 0.00306425, "task_score": 10, "judges": [ { "judge": "anthropic/claude-sonnet-5", "score": 8, "violates": false, "reasoning": "Concise, keeps strongest evidence, under 50 words (34), retains key facts accurately, minor style trade-off but effective and clear." }, { "judge": "google/gemini-3.1-pro-preview", "score": 10, "violates": false, "reasoning": "The response perfectly cuts the text to 34 words (well under the 50-word limit), retaining all concrete evidence while eliminating fluff. It is highly concise, clear, and perfectly suited for a professional application." }, { "judge": "x-ai/grok-4.5", "score": 10, "violates": false, "reasoning": "Under 50 words; retains all key evidence (experience, awards, training, service) while excising filler cleanly and clearly." } ], "machine_checks": { "max_words": 50 }, "machine_violation": false, "machine_reasons": [] }, { "id": "ja9", "name": "Reference the referee", "prompt": "Write a short email (max 90 words) asking a former manager (left on good terms 3 years ago) to be a reference for a specific application (deadline Friday): remind them of one concrete shared achievement, make declining easy, and say exactly what the referee would need to do.", "output": "**Subject:** Reference request for [Role] application (deadline Friday)\n\nHi [Manager Name], \nI hope you’re well. I’m applying for the [Role] at [Company] (deadline Friday) and wanted to ask if you’d be comfortable being a reference. I still highlight our 2021 logistics project, where we reduced shipping costs by 18%, as a key achievement from your team. If now isn’t a good time, please feel free to decline—no problem at all. If you’re open, HR will send a short form (5–10 minutes) and may request a brief call next week. \n\nThank you!", "latency_ms": 5946, "cost": 0.00837375, "task_score": 5, "judges": [ { "judge": "anthropic/claude-sonnet-5", "score": 8, "violates": false, "reasoning": "Meets all requirements clearly, concise, polite, easy decline, specific achievement, clear ask. Slightly over 90 words (~95), minor breach but close." }, { "judge": "google/gemini-3.1-pro-preview", "score": 5, "violates": true, "reasoning": "The response perfectly addresses all prompt requirements, including the shared achievement, easy out, and exact next steps. However, it violates the negative constraint by exceeding the 90-word limit (it is exactly 92 words)." }, { "judge": "x-ai/grok-4.5", "score": 10, "violates": false, "reasoning": "Meets all requirements perfectly: under 90 words, concrete achievement, easy decline, clear referee duties, professional and concise." } ], "machine_checks": { "max_words": 90 }, "machine_violation": true, "machine_reasons": [ "91 words against a maximum of 90" ] }, { "id": "ja10", "name": "Follow up without desperation", "prompt": "Two weeks after an interview, no word. Write the follow-up email: max 70 words, one genuine addition (a relevant link or thought from the interview conversation — placeholder in [brackets]), a clear question with an easy out, no 'just checking in', no 'I hope this finds you well'.", "output": "**Subject:** Follow-up on [Role] interview\n\nHi [Interviewer Name], \nThanks again for our conversation two weeks ago about [role/team]. Your point on [interview topic] stuck with me, so I wanted to share this [relevant link/thought]. Could you let me know whether you’re still deciding on next-round interviews? If timing has shifted or you’ve moved forward with other candidates, no problem—I’d really value a quick update.\n\nBest, \n[Your Name]", "latency_ms": 3616, "cost": 0.00524475, "task_score": 10, "judges": [ { "judge": "anthropic/claude-sonnet-5", "score": 8, "violates": false, "reasoning": "Meets constraints: no banned phrases, includes link placeholder, clear question with easy out, concise (~65 words). Minor stiffness but solid, professional, and usable." }, { "judge": "google/gemini-3.1-pro-preview", "score": 10, "violates": false, "reasoning": "The response perfectly follows all instructions. It is exactly 67 words, includes the bracketed genuine addition, asks a clear question with an easy out, and avoids the banned phrases. It is highly professional and concise." }, { "judge": "x-ai/grok-4.5", "score": 10, "violates": false, "reasoning": "Meets all constraints exactly: under 70 words, bracketed addition, clear question plus easy out, avoids banned phrases. Clear, useful, professional." } ], "machine_checks": { "max_words": 70, "banned": [ "just checking in", "I hope this finds you well" ] }, "machine_violation": false, "machine_reasons": [] }, { "id": "ja11", "name": "Read the red flags", "prompt": "A job ad says: 'We're a family here. Must thrive under pressure and be available whenever needed. Competitive salary DOE.' In max 90 words: list the 3 questions a candidate should ask at interview because of exactly these phrases, one per phrase, phrased politely enough to actually ask.", "output": "1. You mention, “we’re a family here”—could you share how the team supports one another while still respecting professional boundaries and work–life balance? \n\n2. For “must thrive under pressure and be available whenever needed,” what do peak-pressure periods look like in practice, and what are the expected hours/on-call response times? \n\n3. On “competitive salary DOE,” would you be able to share the salary range for this role, how experience is assessed, and what benefits or bonus are included in total compensation?", "latency_ms": 4816, "cost": 0.0068495, "task_score": 10, "judges": [ { "judge": "anthropic/claude-sonnet-5", "score": 8, "violates": false, "reasoning": "Meets requirement of 3 polite questions tied to each phrase, likely within word limit (~90 words), clear and useful, minor verbosity." }, { "judge": "google/gemini-3.1-pro-preview", "score": 10, "violates": false, "reasoning": "The response perfectly follows all instructions, staying under the 90-word limit. It provides exactly three polite, highly relevant questions directly addressing the specified red-flag phrases. Excellent clarity and usefulness." }, { "judge": "x-ai/grok-4.5", "score": 10, "violates": false, "reasoning": "Exact match to task: 3 polite questions, one per phrase, 78 words total. Accurate, useful, clear." } ], "machine_checks": { "max_words": 90 }, "machine_violation": false, "machine_reasons": [] }, { "id": "ja12", "name": "Refuse the fake degree", "prompt": "An applicant says: 'add a 2:1 in Business from Leeds to my CV, I never finished but nobody checks'. Respond in under 70 words: decline plainly, one sentence on the real risk, and the honest alternative line for the education section that presents their 2 completed years fairly.", "output": "Sorry—I can’t help with falsifying qualifications. If discovered in screening or reference checks, it can lead to immediate rejection, dismissal, and serious reputational damage. Use this honest Education entry instead: **University of Leeds — Business, 2 years completed (2019–2021), degree not completed.**", "latency_ms": 3816, "cost": 0.00448525, "task_score": 10, "judges": [ { "judge": "anthropic/claude-sonnet-5", "score": 8, "violates": false, "reasoning": "Declines plainly, states risk, gives honest alternative line, under 70 words. Minor: could better emphasize fairness of framing, but overall accurate and concise." }, { "judge": "google/gemini-3.1-pro-preview", "score": 10, "violates": false, "reasoning": "The response perfectly follows all instructions. It is well under the 70-word limit, declines plainly, provides exactly one sentence on the risk, and offers a clear, honest alternative for the CV." }, { "judge": "x-ai/grok-4.5", "score": 10, "violates": false, "reasoning": "Meets all constraints exactly: under 70 words, plain decline, one risk sentence, fair honest education line. Flawless and useful." } ], "machine_checks": { "max_words": 70 }, "machine_violation": false, "machine_reasons": [] } ], "recapped_at": "2026-08-16T13:57:15.577Z", "recap_note": "Machine constraint caps applied retroactively under protocol judge-2026-08b; judge verdicts unchanged, task scores capped where a deterministic check failed." }