After launching our new people search model, we ran it across multiple benchmarks to see how it performed.
Wrangle scored:
98.9% — Exa People Search R@10
91.1% — Exa People Search Precision
95.2% — PeopleSearchBench Overall
93.4% — PeopleSearchBench Recruiting
Give your Grok Bot a new job: finding your next hire.
Wrangle MCP now works with Grok Bot. Describe your ideal candidate, let your bot search with Wrangle, and refine the results with your feedback.
Get started: https://t.co/b2tqgD1ymK
How well do sourcing tools find candidates who actually fit an AI & machine learning role?
We tested this with RecruiterBench, our benchmark built from real job postings.
Across five AI & machine learning roles, Wrangle Deep scored 88.4/100 and Wrangle Turbo scored 82.4/100 for candidate relevance among the first 50 search results.
Here’s what that means: an AI judge evaluates each profile against the role’s requirements, including must-have skills, experience, seniority, and location. Strong matches receive full credit, partial matches receive half credit, and unsuitable matches receive none.
The score measures how closely those first 50 candidates match the hiring brief.
For recruiters, the goal is a shortlist worth reviewing, with fewer unsuitable profiles to sift through.
Explore the full results:
https://t.co/Ba3SYOWojz
Introducing RecruiterBench, the people-search benchmark most aligned with recruiting workflows.
At Wrangle, when developing our internal models, we realized that all existing people-search benchmarks (that we’ve seen) are flawed:
- They only score 10 candidates
Real recruiting workflows require sourcing through 100 or more profiles to make a hire. Evaluating accuracy across 10 people doesn’t provide an adequate picture of how a tool will perform on a req.
RecruiterBench measures precision over 200 returned candidates for each role, which we believe accurately represents search capabilities and performance.
- The queries are overly simplistic and contrived
Accuracy over “Product managers in Austin” doesn’t matter when you’re looking for someone with a specific skillset.
We built RecruiterBench on 290 complete active reqs on Ashby, spanning 58 unique industries. This also allows us to find unique insights, such as what industries and role types each tool excels or struggles with.
For example, some strong points we identified are:
Wrangle deep on AI infrastructure: p@50 of 60.4%, versus Metaview 39.6%, Wrangle Turbo 52.2%, Exa 21.8%, and Treg 9.6%.
Wrangle deep on food/agriculture: p@50 of 79.4%, versus Wrangle Turbo and Metaview at 56%, Exa 24.6%, and Treg 9.2%.
Juicebox on legal: on the same three roles, p@50 was 82.7%, ahead of Turbo 76.3%, Metaview 65.7%, Exa 55.7%, and Treg 11.7%. Wrangle Deep still led at 93.3%.
Metaview on Advertising/marketing tech: p@50 was 78.2%, leading Wrangle Deep's 77.8%, Wrangle Turbo’s 73.8%, Exa’s 45.6%, and Treg’s 17.6%..
- They evaluate with a single LLM judge, without nuance about must-have, dealbreaker, and nice-to-have criteria
RecruiterBench uses a two-part LLM scoring system to first disqualify candidates based on must-haves, dealbreakers, location mismatches, and over/underqualification. Then we use a second judge to determine relevance and give no/partial/full credit.
We’re excited to continue developing more accurate benchmarks, and to build the best recruiting tool in the world!
Check out RecruiterBench on Github: https://t.co/ACdb6A2wYU.
“Less clicks, more output is essentially what I’m looking for. And that’s what Wrangle does really well.” - Ryan Taft, Talent Lead at Serval
Read the full case study: https://t.co/4YpyAkWuRK
Jev is a phenomenal reranker. At 75% of the previous reranking cost, it improved our main-app search by 30.64% in matches per 200 candidates on searches recruiters actually run.
Most recruiting benchmarks measure precision and accuracy across just 10 candidates. That isn’t representative of real recruiting and sourcing workflows, where you may need to source hundreds of candidates to make a hire.
We’ll release the full benchmark soon.
We're excited to work with @hmartenjoyer, @mathfax , and the rest of the @typesafeai team to continue to bring Jev to more Wrangle features. Stay tuned 👀
Introducing Market Mapping in Wrangle.
“Engineers building payments infrastructure at Stripe.”
“Enterprise AEs who worked at Salesforce between 2018 and 2023.”
Describe the experience you’re looking for. Map where top talent comes from in minutes.
“We’ve switched from using different platforms for sourcing, enrichment, outbound, and inbox to doing it all on Wrangle”
- Henry Xiao, Head of Talent at Judgment Labs
Read the full case study: https://t.co/DL5DZOKVf8
The best talent often comes from a team's network, so we made it easier to make these intros.
Now live in Wrangle: mint a magic link, and anyone in your org can enter their email, sign up in one click, and be brought to a specific network upload page.
Then you can source across your full team's network, see who's connected to any given candidate, and request an intro.
Your next great candidate could start with a prompt.
Describe the person you need, search with Wrangle, and refine the results directly inside ChatGPT.
After launching our new people search model, we ran it across multiple benchmarks to see how it performed.
Wrangle scored:
98.9% — Exa People Search R@10
91.1% — Exa People Search Precision
95.2% — PeopleSearchBench Overall
93.4% — PeopleSearchBench Recruiting
Your edge as a recruiter comes down to your expertise.
Wrangle now remembers the preferences, decisions, and context you build across Search and Agents, and carries them forward into future work.
The more context you give Wrangle, the more it adapts to how your team hires.
When you share a collection, your teammates can now review candidates and leave feedback.
Their input is automatically added to the Project as context, so your team can source together more seamlessly.