Beamery
HireAIScore rates Beamery 66 out of 100 (grade D, Concerning) under rubric v1.0, last reviewed November 14, 2026. Beamery ranks 3 of 86 vendors rated in Candidate sourcing AI, against a category average of 46. That score is drawn from 34 evidence items across 7 rubric criteria for Beamery — 28 with a cited source, 6 recording that nothing was located.
Score breakdown
Beamery, criterion by criterion
Each rubric criterion is scored out of 100 on the cited evidence, then weighted into the total.
Weighted contribution shown to the right of each bar.
Article 11 Technical Documentation
Weight 20% · category avg 49
73
+14.6
Bias Audit Transparency
Weight 18% · category avg 40
85
+15.3
FRIA Support
Weight 15% · category avg 32
40
+6.0
Data Governance Disclosure
Weight 15% · category avg 54
64
+9.6
Human Oversight Design
Weight 12% · category avg 56
64
+7.7
Post-Market Monitoring
Weight 12% · category avg 41
66
+7.9
Customer Documentation
Weight 8% · category avg 57
65
+5.2
Total weighted score
66.3(rounded to 66 for display)
Category avg is the mean raw score on that criterion across the 86 Candidate sourcing AI vendors in scope of this rubric, this one included.
Strongest criterion
Bias Audit Transparency
85raw / 100
Contributes 15.3 to the total. 85 against a 40 category average.
Weakest criterion
FRIA Support
40raw / 100
Contributes 6.0 to the total. 40 against a 32 category average.
Cited evidence
The evidence behind the score
Every score is backed by at least one cited piece of evidence. Open a criterion to see what was found, when, and where.
Download diligence recordEvidence ledger
- Items
- 34
- Documentation
- 21
- Audit report
- 4
- Public statement
- 2
- Product feature
- 1
- Absence
- 6
- With a source URL
- 28 of 34
- Source hostnames
- 5
- Fewest items
- 4
- FRIA Support
These figures measure how thoroughly Beamery was reviewed, not how Beamery performed — an absence row, recording that nothing was located, is counted like any other item.
Article 11 Technical Documentation
5 items73
- DocumentationCaptured Nov 14, 2026
Public 20-page AI Explainability Statement (September 2024, v3) covering purpose, model types, training data, testing, drift and skew monitoring, bias auditing and accuracy limits. It still dates most AI Act obligations to 2 August 2026, which the Digital Omnibus on AI has moved to 2 December 2027.
https://beamery.com/resources/beamery-product/beamery-ai-explainability-statement
- DocumentationCaptured Nov 14, 2026
Public March 2025 model inventory mapping each AI feature to its model (Candidate Match, Entity Recommendation, Entity Reconciliation, Knowledge Graph Embeddings, TalentGPT on Azure OpenAI) and production status.
https://beamery.com/resources/ai-compliance/read-beamerys-ai-functionalities-deep-dive
- DocumentationCaptured Nov 14, 2026
Help-centre FAQ (edited October 2026) publishing the five Match Score factors and their weights: job title 30%, skills 30%, seniority 12.5%, company industry 22.5%, company size 5%.
https://support.beamery.com/hc/en-us/articles/9867879374353-Beamery-AI-Talent-Match-FAQ
- DocumentationCaptured Nov 14, 2026
Vanta trust centre (84 documents, none public) lists AI Model Cards for Candidate Matching, Entity Recommendation, Entity Reconciliation and Knowledge Graph Embeddings, an AI System Lifecycle, AI Scope and AI architecture diagrams, and the ISO 42001 certificate, Statement of Applicability and audit report, all behind Request access.
- Public statementCaptured Nov 14, 2026
ISO/IEC 42001 certification announced August 2025; certification body and scope not named publicly.
Bias Audit Transparency
6 items85
- Audit reportCaptured Nov 14, 2026
Warden AI dashboard for AI Talent Match: monthly audits, latest 5 November 2026, sample size 49,140, 58 Clear and no Consider or Concern results across sex, race/ethnicity, intersectional, age, disability, religion and sexual orientation.
- Audit reportCaptured Nov 14, 2026
Public NYC LL 144 report dated 5 November 2026, data type test data: every reported sex, race and intersectional group has an impact ratio of 1.00; Native American, Pacific Islander and two-or-more-races groups have no records.
- DocumentationCaptured Nov 14, 2026
Customer DPIA guide commits to audits at least annually and to an audit before any material model change is released to customers.
https://beamery.com/resources/compliance/dpia-and-data-transfer-information
- Audit reportCaptured Nov 14, 2026
Original Warden AI audit announcement (August 2024).
https://beamery.com/resources/news/beamery-releases-independent-ai-audit-results-from-warden-ai
- Audit reportCaptured Nov 14, 2026
Earlier audit with Parity (November 2022).
https://beamery.com/resources/news/beamery-completes-ai-audit-for-bias
- AbsenceCaptured Nov 14, 2026
No bias audit for the Ray sourcing agent or TalentGPT, and no employer-published LL 144 summary of Beamery's tool.
Where we looked: trust.warden-ai.com/beamery (AI Talent Match only), beamery.com/resources/ai-compliance, web searches for employer LL 144 notices naming Beamery.
FRIA Support
4 items40
- DocumentationCaptured Nov 14, 2026
Public customer guide for completing a DPIA and transfer impact assessment on Beamery, with AI-specific sections on alternatives to AI, human review, bias mitigation, transparency, accuracy and turning AI off.
https://beamery.com/resources/compliance/dpia-and-data-transfer-information
- DocumentationCaptured Nov 14, 2026
Customisable legitimate interest assessment template for customers.
https://beamery.com/resources/compliance/beamery-legitimate-interest-assessment-for-customers
- DocumentationCaptured Nov 14, 2026
Help article (edited September 2026) with a task-by-task table of where machine learning, generative or agentic AI is used; screening out applicants, interview and hiring decisions are all marked as not using AI.
- AbsenceCaptured Nov 14, 2026
No FRIA template, Article 27 deployer guidance or fundamental-rights analysis, including of skills inference under Article 22 GDPR.
Where we looked: beamery.com/resources/ai-compliance, beamery.com/resources/compliance, trust.beamery.com/resources (no FRIA item), support.beamery.com, web search for Beamery FRIA.
Data Governance Disclosure
5 items64
- DocumentationCaptured Nov 14, 2026
AI page exclusion list (race, ethnicity, gender, nationality, income, sexual orientation, political or religious beliefs are not considered) and a statement that PII and customer data are never used to train Beamery's AI.
- DocumentationCaptured Nov 14, 2026
Explainability Statement says the training set was purchased from a third party with lawful-collection warranties, then de-identified, with protected characteristics and proxies such as postal codes removed.
https://beamery.com/resources/beamery-product/beamery-ai-explainability-statement
- DocumentationCaptured Nov 14, 2026
AI FAQ (edited October 2026) describes the training set as public data of about 2 million profiles and says models are retrained when the taxonomy changes.
https://support.beamery.com/hc/en-us/articles/9869063660689-Beamery-AI-FAQ
- DocumentationCaptured Nov 14, 2026
Retention: customer data kept in production for 30 days after the contract ends, then deleted from backups after a further 90 days.
https://beamery.com/resources/compliance/dpia-and-data-transfer-information
- AbsenceCaptured Nov 14, 2026
No subgroup accuracy, error or calibration metrics and no published validation results; the AI page's 95% skills-label accuracy figure has no stated method.
Where we looked: beamery.com/ai, the Explainability Statement, the March 2025 functionalities document, support.beamery.com AI and match-score articles.
Human Oversight Design
5 items64
- DocumentationCaptured Nov 14, 2026
Customers can turn off Beamery's AI features for all users and candidates.
https://beamery.com/resources/compliance/dpia-and-data-transfer-information
- DocumentationCaptured Nov 14, 2026
Applicant Review (August 2026): no applicant is removed from view by ranking, scoring or filtering and none is auto-rejected; filters are never applied automatically, and an account setting hides match scores.
https://support.beamery.com/hc/en-us/articles/50297099283729-Applicant-Review
- DocumentationCaptured Nov 14, 2026
Users can hover over a match score to see how each factor contributed; a feedback link flags incorrect scores.
https://support.beamery.com/hc/en-us/articles/9671611196305-Beamery-AI-Explained
- DocumentationCaptured Nov 14, 2026
July 2026 release adds an option to match on declared skills only, excluding Beamery-inferred skills.
https://support.beamery.com/hc/en-us/articles/48475237620753-July-2026-Release-Note
- AbsenceCaptured Nov 14, 2026
No documented audit log of AI outputs or of user actions on AI recommendations.
Where we looked: support.beamery.com for audit log, disable AI and hide match score; trust.beamery.com resource titles.
Post-Market Monitoring
5 items66
- Product featureCaptured Nov 14, 2026
Warden AI dashboard serves as a public, monthly post-market bias monitoring surface; latest audit 5 November 2026.
- DocumentationCaptured Nov 14, 2026
September 2026 release note discloses a skills taxonomy revision behind Talent Match (859 new, 2,361 updated, 610 merged, 7 removed skills) and warns that suggested-contact results may vary after the change; the November 2026 note covers a roles taxonomy update.
https://support.beamery.com/hc/en-us/articles/50293938960273-September-2026-Release-Note
- DocumentationCaptured Nov 14, 2026
Describes monitoring of model input data for feature skew and drift.
https://beamery.com/resources/beamery-product/beamery-ai-explainability-statement
- DocumentationCaptured Nov 14, 2026
Public status page (linked from beamery.com/security) with incident history for the CRM, integrations and Workforce Intelligence Suite.
- AbsenceCaptured Nov 14, 2026
No AI-specific incident reporting channel and no model version history.
Where we looked: beamery.com/ai (general whistleblowing address only), beamery.com/security, support.beamery.com release notes January 2024 to November 2026. status.beamery.com has no DNS; the status page is on beamery.co.
Customer Documentation
4 items65
- DocumentationCaptured Nov 14, 2026
Talent Match FAQ explains factor weights, limits (poor matches usually trace to low data quality) and that Beamery's customers are the deployers of Talent Match.
https://support.beamery.com/hc/en-us/articles/9867879374353-Beamery-AI-Talent-Match-FAQ
- DocumentationCaptured Nov 14, 2026
Task-scope table stating which hiring tasks use AI and which do not.
- Public statementCaptured Nov 14, 2026
October 2024: Beamery describes itself as one of the first signatories to the European Commission's AI Pact.
- AbsenceCaptured Nov 14, 2026
No document presented as Article 13 instructions for use and no stated AI Act conformity status.
Where we looked: support.beamery.com AI articles, beamery.com/resources/ai-compliance, trust.beamery.com/resources.
Editorial notes
Company overview
Beamery started as a talent CRM and now positions itself as an AI platform for workforce transformation. Its AI Talent Match model scores candidates against roles from a skills and roles taxonomy, and it also offers workforce planning tools and Ray, an agentic AI assistant that builds sourcing searches. In 2026 it shipped a new Applicant Review experience that ranks applicants with Talent Match but never auto-rejects them.
Regulatory exposure
Talent Match falls under EU AI Act Annex III §4 (obligations from 2 December 2027) and NYC Local Law 144 wherever it influences hiring decisions. The skills-inference layer also raises GDPR profiling questions independent of the AI Act: candidates whose skills are inferred rather than declared have rights under Article 22 GDPR, and Beamery is UK-headquartered, which adds UK GDPR to the same surface. Beamery holds ISO 42001 and runs monthly public bias audits through Warden AI, though on Warden's test data rather than live hiring data.
Path to a higher score
Beamery has one of the strongest monitoring postures in the directory: a public Warden AI dashboard refreshed monthly, release notes that warn customers when model-affecting taxonomy changes ship, and public help articles on factor weights and off switches. The largest gains are (a) publishing the four model cards and AI system documents that sit gated in its trust centre, (b) bias results on live applicant data and coverage of the Ray agent, (c) a FRIA or Article 22 analysis for skills inference, and (d) one consistent account of its training data, which its FAQ and Explainability Statement describe differently.
Regulatory frame
What applies to candidate sourcing AI.
Ranking outputs are inside Annex III §4 when they influence hiring decisions. GDPR Article 22 (automated decision-making) and the GDPR rights around profiling apply independently for EU candidates.
Compare
Beamery against its nearest-scoring peers.
The closest scores in Candidate sourcing AI, side by side.
- Side-by-sideBeamery vs Eightfold AI67D
- Side-by-sideBeamery vs Greenhouse67D
- Side-by-sideBeamery vs SmartRecruiters62D
- Side-by-sideBeamery vs Phenom59F
Ranking
Others rated in Candidate sourcing AI
Ranked 3 of 86 by weighted total under rubric v1.0. The ordering is arithmetic on the rubric and carries no view on which tool suits a given hiring process.
- #1Eightfold AI67D
- #2Greenhouse67D
- #3BeameryThis profile66D
- #4SmartRecruiters62D
- #5Phenom59F
Conflicts of interest
No vendor pays for placement, scoring, or removal. Casework - the consulting firm that operates this directory - provides paid services to some vendors. Any active or recent (within 24 months) commercial relationship is disclosed on the affected vendor profile and the review is reassigned to an independent reviewer. See the full policy on About.
Casework has no commercial relationship with this vendor.