“I want someone who can explain the thinking behind their test strategy.”
The job description and manager priorities inform the assessment.
JL
Jordan Lee
QA Automation Engineer
For your review
Job description fit8.0 / 10
Hiring manager fit7.6 / 10
The evidence behind the fit
Owned API regression coverage and release validation for a customer-facing platform.
View the resume excerpt ↗
“Built and maintained API regression suites and coordinated testing for monthly releases.”
A QUESTION WORTH ASKING
How did you decide what to automate first—and what changed as a result?
Proposed scores. You review before applying.
THE CONVERSATION CHANGES THE PICTURE
This is where you come in.
Your observations add what a resume can’t.
Your screening note
“Built the API tests themselves. Strong ownership. Need to check BDD experience.”
A clearer follow-up
PREPARED FOR YOUR NEXT CONVERSATION
Ask for an example of writing BDD scenarios and how the team used them.
Accept or correct the interpretation. Your judgment stays central.
More than a score.A conversation starter.↗
Built around your way of recruiting.
Role-specific context Evidence you can explore Your judgment at the center
LESS ADMIN. MORE HUMAN.
From first look to a better conversation.
One connected workspace for the details that make a candidate more than a resume.
01
Start with what matters.
Add the role, the must-haves, and the hiring manager’s priorities. Give every assessment the right context.
A clear starting point
02
Find the story in the resume.
Bring in candidates. Explore the supporting evidence, proposed fit, and questions that deserve a closer look.
A more prepared conversation
03
Make your insight count.
Add what you learn in conversations. Review the interpretation and proposed assessment updates.
A fuller picture of the fit
RECRUITING IS PERSONAL. KEEP IT THAT WAY.
You bring the instinct. We bring the context.
The best part of recruiting is connecting with people. blumr helps you prepare, connect the details, and keep your expertise at the heart of the search.
A fresh perspective on your next search. Sign in or make yourself at home.
Enter your existing account details.
A LITTLE MORE CONTEXT
Before you jump in.
What should I start with?+
Start with a job description and the hiring manager’s priorities, then add candidate resumes. The Learn blumr walkthrough can guide you through the workflow with sample candidates.
Does AI make the hiring decisions?+
You make the decisions. blumr prepares analysis and highlights evidence and open questions. You review proposed assessment changes and can accept or correct feedback interpretations.
Can I explore before creating an account?+
Yes. The interactive preview at the top uses illustrative sample data. Switch between the role, candidate, and feedback views to get a feel for the experience.
✓
Welcome to blumr
Your account and private workspace have been created successfully.
Reset your password
Your username is the work email used to create your account. Enter it below and we’ll send a secure reset link.
Choose a new password
Enter a new password for your blumr account.
Saved
Private workspace
Choose a job to open
Select the search you want to work on. Each job keeps its candidates, evaluation criteria, benchmarks, feedback, and interview outcomes separate.
Your workspace data is securely loaded from your blumr account.
Search overview
Review the shortlist and the latest assessment updates.
Automatic assessmentsChecking for updates…Review updates ⌄
Saved criteria and approved shared preferences automatically review this job’s candidates. Candidate feedback updates that person’s assessment. Review each AI proposal before applying it.
Search setup
Build stronger recommendations step by step
0 of 6 complete⌄
Complete these steps as your search develops. blumr will keep track for you.
★
Top Candidate
—
Loading active candidates…
↗
Average Fit Score
—
Across active candidates
◎
Candidates Reviewed
0
Resume / portfolio assessments
✓
Interview Recommendations
0
No interview recommendations yet
Candidate Rankings
Rank
Candidate
JD Fit
Manager Fit
Recommendation
Evaluation Criteria Weighting
What the Hiring Manager Values Most
Benchmark Comparison
Recently Added / Updated
Pipeline Overview
Active search
Fit Score Distribution
Manager fit
Recent Activity
Top Candidates (by Fit Score)
Search Summary
Candidates
Find a candidate, review the assessment, or upload a resume.
Add candidate
Upload a resume to create a candidate and prepare a screening brief automatically.
Add candidate details manually
Rankings
JD Fit is the objective baseline. Manager Fit is calibrated using benchmarks and hiring-manager feedback.
Rank
Candidate
JD Fit
Manager Fit
Confidence
Recommendation
Primary signal
Benchmarks
Add or remove reference candidates for the active job. Removing benchmark status does not delete the candidate.
Criteria and screening questions
Save a job to prepare criteria automatically.
Evaluation criteria
Adjust weighting as new hiring-manager feedback comes in.
Total: 100%
Hiring manager feedback
Add feedback and preserve it as context for future evaluations.
Saved feedback
Manager preference profile
Review what blumr should learn for this search. Nothing proposed changes rankings until you approve it.
Low confidence
Interview outcomes
Track interview results and preserve what the team liked, disliked, and decided.
Saved interview outcomes
Candidate pipeline
Move candidates forward and record what happens next.
Hiring insights
Learn what the manager values, then decide whether those insights should update candidate evaluations.
1
Analyze hiring patterns
Review criteria, approved manager preferences, and interview decisions.
→
2
Update candidate evaluations
Create reviewable score proposals using the latest analyzed evidence.
Hiring evidence analysis
blumr checks the evidence and explains what it suggests. You approve every change.
Evidence checksAI interpretationYour approval
Not analyzed
Add manager feedback or interview activity, then select Analyze Hiring Patterns above.
Candidate evaluation updates
Proposed changes use the latest analyzed patterns and current approved evidence. Scores remain unchanged until you apply them.
Complete Step 1 above, then update candidate evaluations to create score proposals.
What this manager actually hires
Learning
Criterion / signal
JD priority
Positive outcomes
Passes
Observed importance
Recurring decision patterns
Calibration confidence
Resume vs. interview gaps
Job health
Recommended calibration changes
Candidate comparison
Select up to four candidates and compare their fit, risks, and interview evidence side by side.
⇄
Build a comparison
Choose two to four candidates for the clearest side-by-side view.
01
02
03
04
◎
Your comparison will appear here
Select candidates above to compare fit, strengths, risks, and interview evidence.
New to blumr?
Take a five-minute tour, or start your first job below.
Jobs
Create multiple searches, store each job description and manager feedback, and switch the dashboard between them.
Your searches
Add Job
Loading practice progress…
Learn blumr
Your first search, with a little guidance.
Work through a sample job, review the evidence, and prepare a shortlist you can explain.
5 steps · at your pace
Fictional data. Practice creates no real jobs or candidates and runs no AI analysis.
YOUR PRACTICE SEARCH
QA Engineer
JL
Jordan Lee
Ready to review
AR
Alex Rivera
Ready to review
?Learn one action at a time, right where the work happens.
01Set up a search
02Add resumes
03Review evidence
04Capture feedback
05Build a shortlist
Just need a quick answer?
Why are there two fit scores?
JD Fit measures alignment with the job description. Manager Fit adds the manager’s priorities and approved feedback. Open the supporting evidence before relying on either score. Missing evidence gives you a question to ask.
What happens after I save a note?
A short note on the candidate saves automatically after a pause. Your original wording is kept. Review the AI’s interpretation and correct it if needed. A fact about one candidate belongs to that candidate; a reusable manager preference needs its own review. Check the proposed changes before approving a new assessment. An explicit Move forward or Pass outcome can also adjust Manager Fit. Approving a reusable preference can affect matching candidates across the job.
One resume failed. Do I start over?
No. Upload resumes on Candidates accepts several files, and each gets its own result. Keep the successful uploads and retry the failed file. If it still cannot be read, try a supported PDF, DOCX, or TXT copy with selectable text.
How do I close or reopen a search?
Use Close job from Jobs and record the reason. Marking a candidate Hired also closes the job. Closed searches keep their history; use the Closed filter to find and reopen one.
When should I use a reusable manager preference?
Use a reusable preference when the feedback describes what this manager wants across the search. Keep observations about one person candidate-specific. Review the proposed preference before approving it; only matching evidence should affect another candidate.
◇Practice mode · fictional data
Step 1 of 5
Give the search a clear starting point.
A job brief tells blumr what to look for.
?
We’ve filled in a sample role. Notice how the manager’s priority is more specific than the job title.
Step 2 of 5
Add resumes in one batch.
QA Engineer
?
Each resume gets its own status. Let’s try a batch with one file that needs a retry.
+
Three resumes, one upload
In your workspace, choose PDF, DOCX, or TXT files.
▤
Jordan_Lee.pdf
Candidate created
Ready to review
▤
Alex_Rivera.docx
Candidate created
Ready to review
▤
Morgan_Chen.pdf
File could not be read
Needs retry
2 ready · 1 needs attention
Step 3 of 5
Check the evidence behind the score.
Scores start the review. Source evidence makes it useful.
?
Open the resume excerpt. For Jordan, automation ownership is a question to ask, not a reason to assume they lack the skill.
JL
Jordan Lee
Candidate 1 of 3
Example assessment
JD Fit
7.8 / 10
Evidence against the job description
Manager Fit
6.2 / 10
Evidence against manager priorities
What’s supported
Regression testing and clear defect documentation.
What to ask next
Did Jordan personally build and maintain automated tests?
“Wrote regression test cases, documented reproducible defects, and partnered with the automation team to extend regression coverage.”
Open the source before approving.
✓3 assessments reviewed
You’ve got the review rhythm.
Check the evidence, approve the assessment, then move to the next candidate. Approval accepts an assessment; it doesn’t make a hiring decision.
Step 4 of 5
Add what you learned.
A short screening note can fill a gap in the resume.
?
Your practice call answered Jordan’s automation question. Save the note, then check that the interpretation preserves its meaning.
JL
Jordan Lee
✓Original note saved
Review the interpretation
Jordan only
Jordan has hands-on experience building and maintaining test automation.
Why is this only about Jordan?
The note adds evidence about this candidate. It does not change what the hiring manager wants from everyone else.
Example reassessment · awaiting approval
Here’s what the new evidence changes.
Jordan’s automation ownership is now supported by the screening note.
JD Fit
7.8 → 8.4
Manager Fit
6.2 → 7.9
These example changes need your approval.
✓Feedback reviewed and approved
Jordan’s assessment now includes the new evidence. The original note remains available.
Step 5 of 5
Make the shortlist easy to explain.
Turn the reviewed evidence into a useful candidate summary.
?
Review the draft as you would before sharing it. Keep the proof, keep the open question, and make the wording your own.
✓
Practice complete
You’re ready for your first search.
You’ve worked through the decisions that make a shortlist useful.
✓Set the role and manager priorities
✓Added a batch of resumes
✓Checked sources before approving assessments
✓Reviewed the meaning of a screening note
✓Saved a candidate submission draft
Next: use your own job.
Follow the same workflow in your workspace. Return to Learn blumr whenever you need a reminder.
Settings
Make blumr work the way you do.
Save status
No sync confirmed yet in this tab.
If a save fails, keep this tab open and use Retry in the top bar. Candidate workspace notes and summary edits are protected from accidental reload while unsaved.
Appearance
Make it feel like your workspace. Choose a look below—it applies instantly and is remembered on this device.
blumr Mint
10 shades of green · 7 dark themes · 3 light themes. Success, warning, and error colors stay consistent.
Notifications
Help & onboarding
Practice the workflow or find a quick answer.
Report a problem
Describe what happened and what you expected. Reports go privately to the blumr administrator. Include only the information needed to explain the problem.
Your problem reports
Your activity
Your account’s activity over the last 30 days.
Loading activity…
Account Security
Create or replace the password for your signed-in account.
Private account
Ends sign-in sessions. Existing access tokens can remain valid until they expire.
Your data
Export your profile, preferences, accessible workspace records, assessments, and notes as JSON. Original resumes are available as individual downloads after export.
Delete account
This permanently removes your account, private workspaces, candidates, resumes, and preferences. Download your data first if you need a copy. Shared workspace ownership must be resolved before deletion.
Admin tools
Technical setup and deployment resources.
Usage analytics
Account activity, saved work, and completed AI operations. Refresh to load the latest recorded totals.
Loading usage…
Accounts
—
Active · 7 days
—
Active · 30 days
—
Sessions · 30 days
—
Events · 30 days
—
Tester Activity · All Time
Admin only
Tester
Last activity
Sessions
Events
Jobs created
Candidates added
AI completed
Notes saved
Outcomes saved
Loading usage…
Saved-work totals count new records once; edits and retries do not add another record. AI totals include resume analysis, reassessments, criteria refinement, screening, and feedback interpretation. Failed processing is listed separately. Sessions count workspace loads, not time spent. Background processing does not mark a user active.
Feature Activity · Last 30 Days
AI quality review
Recruiting scenario benchmark
Synthetic scenarios
Test how the AI responds when new evidence arrives. Twenty scenarios make 40 AI requests using your existing connection and incur normal analysis usage. No live jobs or candidates are changed.
Expectations are provisional: a decrease of at least 0.3 for weaker ownership evidence, an increase of at least 0.3 for stronger discovery evidence, and a change within ±0.5 when no new capability evidence exists. Your review determines whether the reasoning is sound.
Reports and review notes stay in this tab until exported. Export before refreshing or starting another run. Synthetic cases are inspired by recruiting patterns, not reproductions of historical candidate records.
Ready. Review criteria alongside each result.
Scores, evidence, and activity
JD Fit
Manager Fit
Why they fit
Screening Questions
Concerns / Questions
Primary signal
Activity Timeline
Decision support
Submission Readiness
A requirement-by-requirement view of what is proven, missing, or still needs to be screened.
YOUR NEXT SEARCH STARTS HERE