The core extracted values were correct on the clean resume, including Rugved Nichite, rugved.nchite@email.com, 9876500000, Vasind, the work histories, the education details, and the 2.9-year total experience figure.
What was measured
Accuracy
Are extracted values correct and complete?
decisive for this rankingtransformation
Correct and complete values are the essence of resume parsing, so this directly determines whether the tool succeeds. (3 of 3 judges)
What was given, what came back
Test input: Clean single-column resume — Rugved Nichite · pdf · group: resume-parsing
Input — what we sent
Clean single-column resume — Rugved Nichite
A professionally structured single-column resume for Rugved Nichite, used as the baseline input for parser accuracy across standard resume fields.
Why this input is hard
- · baseline field extraction
- · contact info accuracy
- · work experience parsing
- · education and CGPA extraction
- · skills and certifications extraction
Output — unretouched
Also checked on this input — same tool, 4 other criteria
Field coverage✓ WorkedOn the clean resume, the parser returned the six core fields expected for resume coverage: name, email, phone, work experience, education, and skills.Noise in output✗ FailedThe skills field collapsed into one unspaced concatenated string, so the output was not machine-readable as separate skill items.Noise in output✗ FailedThe parser preserved raw bullet/number formatting in responsibilities, leaking duplicated bullet characters into the extracted text.Output quality◐ MixedThe clean-resume output was moderately usable: the basics were correct, but certifications and GPA were missing and the responsibilities/skills sections still showed formatting artifacts.
Provenance
- Observation
- b72526b7-24ba-4e63-aba9-f8bd88bc7f6e
- Evidence run
- cbbef4db-964c-49fa-a57f-a2977822bdfc
- Study
- Parse resumes into structured data using an API
- Research task
- 86b9jm30n
- Tested at
- not recorded
- Source
- first-party
- Evidence state
- verified
- Proof shown
- input + output shown
- Cost / latency
- not captured
- Repeat run
- not captured
- Tester
- not captured
The last three rows are honest blanks, not placeholders — our capture has no field for them yet.
Query this
get_evidence({
tool: "skima-ai",
scenario: "resume-parsing"
})MCP · mcp.aidemos.com/api/mcp
Free with attribution.
Same input, same check — 9 other tools
measured on Accuracy
Affinda◐ MixedIt split one LinkedIn profile into two website fields and dropped the /in/ path segment, so the URL was not reconstructed as a single complete value.Airparser◐ MixedSilently truncates a compound job title, returning "AI Research Analyst" and dropping "& Software Developer".CVParserPro✗ FailedThe experience field was badly inflated: the profile header showed 11 years, while the report says the candidate actually had only about 2–3 years.Extracta.ai✓ WorkedCorrectly extracts the core personal and work-history values from the clean resume, including both jobs, the education entry, two certifications, and the full 15-skill list.Hireability✗ FailedName accuracy failed: GivenName was split as 'Dev' and FamilyName as 'loper', producing FormattedName 'Dev loper' instead of the candidate's real name.HrFlow◐ MixedOn the clean resume, it still misses a listed task bullet ('Evaluated 10+ AI/ML APIs...') and does not return CGPA.LlamaParse◐ MixedThe clean resume education GPA was captured, but as the string "CGPA: 8.2 / 10" instead of a clean normalized numeric value.OpenResume✗ FailedFails to extract the candidate name even when it is present in the source; the Name field is left empty.Parseur✓ WorkedOn the baseline resume, the identity/contact values were extracted correctly, including the candidate name, title, location, phone number, and LinkedIn URL.
This evidence is published in
Real inputs and real outputs, no retouching · every cell queryable via API & MCP · aidemos.com