Available today
Pipeline builder
Compose resume, questionnaire, assessment and other module types into one ordered pipeline. Pin a version so in-flight candidates are unaffected by edits.
For engineering managers
Run the same pipeline for every candidate, and see the evidence behind every score.
Pin the rounds, invite candidates, and review a shortlist where each dimension carries reasoning, evidence and confidence.
Available today
The pipeline your team already runs.
These parts of boson run end to end today.
Available today
Ranked shortlist
See candidates ordered by a per-job score with the dimension breakdown behind it, including how much of the result comes from stub or low-confidence rounds.
Available today
Evidence-backed scorecards
Each dimension carries reasoning, evidence, confidence and provenance. Open a score to see the quote, answer or event that produced it.
Available today
Resume review with receipts
A resume is graded against the job description across relevance, technical depth, communication and role fit. Quotes are verified against the source text before they are shown.
Available today
Questionnaire
Written questions in the candidate portal. Choice answers and required-field checks are deterministic; free text is graded with structured LLM output, evidence and confidence.
Available today
MCQ and SQL assessment
Multiple-choice and SQL questions with deterministic scoring. The candidate SQL preview runs read-only against a throwaway in-memory database.
Available today
Candidate portal
Candidates get one clear next step, a visible deadline, autosave, reconnect handling and plain-language status. Consent is collected before AI evaluation starts.
Available today
Consent and audit controls
Separate consent per recording type, access logs, retention dates and a deletion-request path in the candidate portal. Audio and editor replay are only captured after their own consent; webcam is not offered today.
Available today
Yoga, the recruiter assistant
Ask Yoga to draft a job, suggest a form or pipeline, explain a scorecard or compare candidates. Reads run directly; every write arrives as a proposal the recruiter applies.
How it works
From pipeline definition to a score you can explain.
The workflow uses the rounds that are available today. The roadmap list stays separate.
- 01
Define the rounds
Choose the modules that predict performance for the role and put them in order.
The pipeline builder composes resume, questionnaire, assessment and other module types into one ordered pipeline. Use the rounds your team can review with evidence.
- Order resume, questionnaire and assessment rounds
- Pick the modules that fit the role
- Keep one pipeline per role
- 02
Pin the version before you invite
Pin a version so in-flight candidates are unaffected by edits.
Editing a pipeline does not have to change an application already in flight. Save the next version and keep candidates on the rounds they started.
- Pin a pipeline version
- Keep in-flight candidates on their version
- Prepare the next version separately
- 03
Invite and collect candidate answers
Invite one candidate to a single module, or share a public link with optional expiry and a use cap.
Candidates get one clear next step, a visible deadline, autosave, reconnect handling and plain-language status. Consent is collected before AI evaluation starts.
- Use a single-use link for one candidate
- Use a rate-limited public link for a wider invite
- Collect consent before AI evaluation starts
- 04
Review the shortlist
See candidates ordered by a per-job score with the dimension breakdown behind it.
The shortlist includes how much of each result comes from stub or low-confidence rounds, so your team knows which scores need a closer look.
- Order candidates by a per-job score
- See the dimensions behind each score
- Spot stub and low-confidence rounds
- 05
Explain a score
Open a score to see the reasoning, evidence, confidence and provenance behind it.
Each dimension carries reasoning, evidence, confidence and provenance. Yoga can explain a scorecard when a teammate asks how a result was produced.
- Open the quote, answer or event behind a score
- Check confidence and provenance
- Ask Yoga to explain a scorecard
Roadmap
Coming later for the rest of the pipeline.
These modules are still being built. They are not part of the workflow above.
Coming later
Coding with sandbox and replay
A timed coding round with test-case scoring, editor replay, soft anti-cheat flags and process grading. The sandbox needs Judge0 and is not live yet.
Coming later
System design
Open-ended design answers with a text and/or diagram deliverable. The grader is still a stub, so no system design score is trusted yet.
Coming later
Command center
Funnel and drop-off per module, score distributions, time in stage, cost per candidate and a needs-review queue. In progress.
Coming later
Calibration and fairness reports
Human-labelled calibration sets, agreement metrics, re-grading on rubric changes, and score distribution reports where lawful group data is available. Coming soon.
Review the rounds that already run end to end.
Use the pipeline builder, ranked shortlist, evidence-backed scorecards, resume, questionnaire, assessment, candidate portal, consent controls and Yoga today.