{"slug": "jack-and-jill-cut-candidate-screening-costs-by-88-with-jev", "title": "Jack and Jill Cut Candidate Screening Costs by 88% with Jev", "summary": "Jack & Jill, an AI-agent-brokered talent marketplace that has worked with over 5,000 companies and more than 400,000 professionals and raised $60M, replaced Google Gemini 3.1 Flash Lite with TypeSafe's Jev model for 100% of calls in a key candidate-matching stage, cutting cost per 1,000 candidates scored from $0.755 to $0.092 — an 88% reduction — and projecting $500K in annual savings. In a test across 150 live roles, Jev kept 94.6% of hiring-manager-requested candidates versus 93.9% for Gemini, a difference the company said was not statistically significant (90% confidence interval of −0.4 to +1.8), and posted an AUC of 0.933 versus 92.4% for Gemini. Median screening time fell from 20.3 seconds to 10.3 seconds per candidate, and Jack & Jill rolled out Jev in production within 10 days of its first test, citing \"roughly 50% faster task completion time.", "body_md": "# How Jack & Jill Cut Candidate Screening Costs By 88% With Jev\n\nWe saved ~half a million dollars last week with Jev.\n\nAnd our products got much better at the same time.\n\nMatthew Wilson, Co-founder/CEO @ Jack & Jill\n\n### Background\n\nJack & Jill is a talent marketplace brokered by AI agents. Jack is a career agent for top professionals and Jill is an agent for hiring teams. When Jack and Jill agree there is a mutual match, they introduce the candidate and the company directly, skipping the application altogether. To date, they’ve worked with over 5,000 companies and more than 400,000 professionals working in startups - and raised a total of $60M in the process.\n\nWithin 10 days of their first test, Jack & Jill had fully rolled out Jev in production - replacing Gemini 3.1 Flash Lite for 100% of calls for a key stage in their candidate matching pipeline.\n\nCompared to Gemini, Jev maintained the same level of accuracy, with 88% lower cost and roughly 50% faster task completion time.\n\nCost savings are projected to be $500K per year - and more importantly, new use cases have been unlocked across the product, such as showing candidates matched jobs immediately after signup.\n\n### Impact\n\n| Result | Details | \n|---|---|\n| ↓ $500K/year | Absolute projected savings | \n| ↓ 88% cost | `BEFORE \\| $0.755` with Google Gemini`AFTER  \\| $0.092` with TypeSafe Jev Cost per 1,000 candidates scored | \n| 2x speed | `BEFORE \\| 20.3 sec`` AFTER  \\| 10.3 sec` Median time to screen each candidate  | \n| No loss in quality | `BEFORE \\| 93.9%` Baseline`AFTER  \\| 94.6%` Candidates requested by hiring managers | \n\n## The challenge: expensive to screen a large volume of candidates with Gemini\n\n#### Before Jev\n\n1. **Candidate pool.** Every day, 6,000 candidates**for every open role** are pulled from the database\n2. **Initial scoring.** A low-cost model (Gemini 3.1 Flash Lite) scores each candidate and creates a shortlist of the top 2,000 candidates\n3. **Close review.** A higher-power model assesses the top 2,000 candidates, ending with a list of a few hundred candidates for the hiring manager\n\nEven with a “low-cost” model like Gemini Flash Lite, costs were adding up fast, due to the \"Initial scoring\" step that assessed 6,000 candidates for every open role.\n\n## Results with Jev: Similar quality to Gemini at 88% lower cost\n\n#### With Jev\n\nBefore switching, Jack & Jill tested both Gemini and Jev models on 150 live roles, and the same candidates for each role.\n\nFor each role, they included:\n\n- every past “winner”: a candidate a hiring manager later asked to meet\n- a sample of remaining candidates (weighted to accurately represent the original 6,000 candidates retrieved for each role)\n\nThe test counted how many winners each model kept on the shortlist. For example, a score of 95% means that for every 100 candidates that the hiring manager asked to be introduced to, the model correctly kept 95 of them in its shortlist.\n\n*Ultimately, Jack & Jill found that Jev was able to maintain similar quality to Gemini, with 88% lower cost and half the median time to complete scoring.*\n\nOverall, the team expects to save more than $500K based on projected volumes over the next 12 months.\n\n### Quality: Slightly higher than Gemini (Recall and AUC)\n\nJev kept slightly more winning candidates than Gemini (+0.7 percentage points), although the difference was not statistically significant (90% confidence interval of −0.4 to +1.8)\n\nJev also performed slightly better on AUC, a measure of how often a model ranks a winner above a randomly chosen candidate with no recorded hiring manager selection. 1.0 is perfect and 0.5 is a coin toss. Jev’s score of 0.933 means it ranked the winner higher in approximately 93.3% of those pairings, compared with 92.4% for Gemini — aligning better with the recorded hiring manager selections.\n\n| **Measure** | **Gemini 3.1 Flash Lite** | **Jev** | \n|---|---|---|\n| % winners kept in the Initial Scoring of 2,000 candidates | 93.9% | 94.6% | \n| Ranking quality (AUC) | 0.924 | 0.933 | \n\n### Speed: Roughly twice as fast\n\nMedian time to screen candidates dropped by 50% with Jev compared with Gemini.\n\n| **Measure** | **Gemini 3.1 Flash Lite** | **Jev** | \n|---|---|---|\n| Median time to screen all candidates for a role | 20.3s | 10.3s | \n\n### Cost: 88% lower token cost, $500K in projected savings\n\nJev's input price is a sixth of Gemini's and its output is free. As a result, the cost per 1,000 candidates with Jev is just 9 cents - compared with 76 cents with Gemini.\n\nJust at their current volume of roles, Jack & Jill will save $265K a year. Factoring in expected growth, the team expects to save **$500K over the next twelve months.**\n\n| **Measure** | **Gemini 3.1 Flash Lite** | **Jev** | \n|---|---|---|\n| List price per million tokens | $0.25 in, $1.50 out | $0.042 in, output free | \n| Cost per 1,000 candidates screened | $0.755 | $0.092 | \n\n## Improving user experience by unlocking new features\n\nJev’s usage has expanded across more than 15 workflows across Jack & Jill's product, ultimately powering a better product experience for users. For instance:\n\n- **Matched roles at sign-up.** Every new candidate is checked against every suitable open role straight after onboarding, for under $0.10 per candidate.\n- **A near-instant profile check.** The check candidates wait on after onboarding dropped from about 3.6 seconds on Gemini to about 0.2 seconds, with better accuracy.\n- **Checks across the product,** including the CV editor, company classification and marketplace filtering.\n\nSavings from using Jev instead of Gemini in these additional features are excluded from the projections above.\n\n## Conclusion\n\nWithin just 10 days, Jack & Jill used Jev as a “drop-in” replacement for Gemini 3.1 Flash Lite - achieving comparable results on quality, with 88% lower cost and 50% median completion time.\n\nIn addition, adopting Jev has unlocked new use cases that will power better user experience throughout Jack & Jill’s product, resulting in faster / better job matches for both hiring managers and professionals considering their next move.\n\n###### Methodology: Quality figures come from an offline test on live roles. The $500K figure is a projection based on expected growth in search volume (cost savings based on current volume = $265K). Gemini list prices are as published by Google at the time of measurement.\n\n∵ ⩆\n\n⩆ ∵", "url": "https://wpnews.pro/news/jack-and-jill-cut-candidate-screening-costs-by-88-with-jev", "canonical_source": "https://typesafe.ai/blog/jack-jill-jev-case-study", "published_at": "2026-11-11 06:54:50+00:00", "updated_at": "2026-11-11 07:23:21.399340+00:00", "lang": "en", "topics": ["artificial-intelligence", "ai-agents", "ai-products", "ai-tools"], "entities": ["Jack & Jill", "Jev", "TypeSafe", "Google Gemini", "Gemini 3.1 Flash Lite", "Matthew Wilson"], "also_reported_by": [], "alternates": {"html": "https://wpnews.pro/news/jack-and-jill-cut-candidate-screening-costs-by-88-with-jev", "markdown": "https://wpnews.pro/news/jack-and-jill-cut-candidate-screening-costs-by-88-with-jev.md", "text": "https://wpnews.pro/news/jack-and-jill-cut-candidate-screening-costs-by-88-with-jev.txt", "jsonld": "https://wpnews.pro/news/jack-and-jill-cut-candidate-screening-costs-by-88-with-jev.jsonld"}}