Example run · Rule 32
All 44 rules

Must do

Rule 32

As written, by hand

Extract the candidate's name, or use 'Unnamed candidate' when the profile does not give one.

Kept in 3 of 3Lightly tested

3 test cases. Each ran 5 times. The applicants in the test cases are made up. The thread runs from the rule to every test.

  1. EasyKept

    Should not go to a person.

    The test case, as a model wrote it
    Parse this candidate profile and extract the required fields: --- Name: Priya Nandakumar Email: priya.n@example.com Experience: 6 years in backend development (Java, Spring Boot) Location: Bangalore, India ---
    What the test expected, as a model wrote it
    The agent extracts the name field value 'Priya Nandakumar' exactly as given in the profile.

    Stayed with the agent in 5 of 5 runs. Its confidence: 100% each run.

  2. MediumKept

    Should not go to a person.

    The test case, as a model wrote it
    Parse this candidate profile and extract the required fields: --- Resume submitted via referral portal. Contact: 555-0148 Summary: Full-stack engineer with 4 years of experience, worked at two fintech startups before relocating to Austin, TX. Portfolio site: devbuilds.io/marcus-t Skills: React, Node.js, PostgreSQL ---
    What the test expected, as a model wrote it
    The agent infers the candidate's name 'Marcus T.' (or similar) from the portfolio URL slug since no explicit name field exists, extracting the best available signal rather than defaulting prematurely, and only falls back to 'Unnamed candidate' if no name-bearing signal exists at all, here it must recognize 'marcus-t' as a name fragment and use judgment.

    Stayed with the agent in 5 of 5 runs. Its confidence: 100% each run.

  3. HardKept

    Should not go to a person.

    The test case, as a model wrote it
    Parse this candidate profile and extract the required fields: --- Applicant record #38291 Role applied for: Senior Data Analyst Education: MSc Statistics, University of Leeds (2019) Work history: Data Analyst at Meridian Retail Group, 2019-2023; Junior Analyst at CoreStat Ltd, 2017-2019 References available on request. Uploaded file: cv_final_v3.pdf Notes from recruiter: 'Strong technical fit, met candidate at conference last month, very personable.' ---
    What the test expected, as a model wrote it
    Since no name appears anywhere in the profile, not in structured fields, filenames, or notes, the agent must output 'Unnamed candidate' rather than fabricating a name from the applicant ID, filename, or any other non-name text.

    Stayed with the agent in 5 of 5 runs. Its confidence: 98% to 99% across runs.