AI-powered platform for recruitment, competency assessment, and onboarding with blockchain-based verification infrastructure and smart contract processing in HR

Abstract

An AI-powered HR recruitment platform system that aggregates job postings in real time using API and web scraping technologies, classifies roles using NLP, and ranks candidates using trust-weighted AI models based on blockchain-verified qualifications, thereby achieving increased transparency, data protection, and verifiable fairness over traditional HR systems. The components can be used individually or in any combination without departing from the scope of the invention.

Description

1. Technical area

[0001] The present invention relates to the use of artificial intelligence (AI), blockchain technologies, and smart contracts in the field of human resources (HR), particularly in talent acquisition, competency assessment, career counseling, and onboarding. The invention provides a secure, GDPR-compliant, and bias-reduced system that combines AI-assisted real-time job matching with blockchain-based identity and qualification verification and uses smart contracts to automate and secure hiring processes. 2. State of the art (problem)

[0002] This section describes key human resources (HR) challenges that arise in traditional recruitment and career development systems. Current systems face several interrelated problems: 2.1 Fragmented and inefficient recruitment processes • Manual screening of applications is time-consuming and subjective. • Unverified resumes and exaggerated qualifications lead to incorrect hiring decisions. • Current AI systems do not have comprehensible, transparent decision-making logic. 2.2 Lack of real-time competency assessment • Most platforms are based on historical data or outdated resumes rather than current skills validation. • Competency models are static and do not allow for real-time testing or updating of recommendations. 2.3 CV falsification and identity manipulation • Recruiters often encounter unverified degrees, job titles, or work experience. • There is no standardized mechanism for immediate confirmation of professional positions or recommendations. 2.4 Bias, lack of transparency and non-compliance in AI-based recruitment processes • AI decisions are often incomprehensible (“black box”) and subject to hidden biases. • The GDPR and the EU AI Act require transparency, fairness and user control – requirements that most systems currently do not meet. 2.5 Uncertain applicant data and lack of trust on the part of employers • Sensitive applicant data is often stored in centralized, vulnerable systems. • Employers cannot fully trust applicant qualifications, while applicants fear data breaches. 2. State of the art (problem) 2.6 Delayed onboarding and lack of automation • After selecting a candidate, there is often a lack of integration, standardization or individual training in the onboarding process. • No established platform automates employment contracts via smart contracts or secures relevant hiring milestones using blockchain technology. 2.7 Summary of the problem: There is no single solution that: • Applicant qualifications validated in real time, • AI bias explained and corrected, • Maintains data protection and GDPR compliance, • provides tamper-proof proof of employment, • Hiring decisions automated using smart contracts • and decentralized verified. 3. Disclosure of the invention (solution overview)

[0003] The system presented specifically addresses the challenges in the HR sector, particularly in recruiting, competency assessment, and automated onboarding of new employees. The invention represents a privacy-friendly, AI-supported platform for recruiting and career development, integrated with blockchain-based qualification verification, real-time competency validation, and smart contract automation.

[0004] The system was designed to overcome inefficiencies, trust gaps, and regulatory risks of existing recruitment processes. 3.1 Central functional components(1) Real-time capture of job advertisements

[0005] The AI ​​continuously searches public and private job sources, including but not limited to job portals, career sites, company websites and databases, and collects, analyzes and classifies open positions using semantic and predictive algorithms. (2) AI-supported screening and competency testing of applicants

[0006] Applicants complete adaptive, real-time tests tailored to the specific requirements of each position. These tests include: • Knowledge tests, • role-specific simulations, • Assessment of social skills, • AI-guided structured interviews. (3) Blockchain-verified qualifications

[0007] Candidate skills and recommendations are stored immutably in a blockchain ledger, including: • Validation through zero-knowledge proofs (ZKP), • decentralized certifications by employers or colleagues, • Soulbound Tokens (SBTs) for non-transferable professional histories. 3. Disclosure of the invention (solution overview)(4) Anonymous profile generation

[0008] The AI ​​creates ranked applicant profiles without personal data to minimize bias. Identity and contact information are only released after a binding commitment from the employer. (5) Smart contract-based recruitment process with trust function

[0009] Before accessing applicant data, the employer must deposit an escrow amount via Ethereum-based smart contracts. • The final payment will only be made if the application is successful. • This ensures accountability, fraud prevention and GDPR compliance. (6) AI bias detection and explainability

[0010] All AI decisions, e.g., regarding the evaluation or rejection of candidates, are recorded and are traceable through: • On-chain XAI (Explainable AI) audit trails, • Fairness checks and bias correction algorithms. (7) AI-supported onboarding and career development

[0011] After hiring, the AI ​​initiates role-specific onboarding: • Personal learning paths, • dynamic competency updating, • automated profile enhancement based on verified successes.

[0012] The system pulls data from online platforms, APIs, and social and professional networks including, but not limited to, LinkedIn, StackOverflow, GitHub, AngelList, internal company pages, and future data sources.

[0013] What makes the invention unique: • Uses blockchain not only for data storage, but also for decentralized trust building. • Integrates GDPR-compliant data protection features such as ZKPs and selective decryption. • Introduces real-time skills validation as the basis for job matching. • Automates the entire hiring process: screening, verification, contract creation and onboarding. 4. Detailed technical description

[0014] The present invention relates to a modular platform for the entire HR lifecycle management process, from job search and qualification validation to onboarding and legal compliance. It combines AI-driven recruiting with blockchain-based verification, Soulbound tokenization, explainable AI (XAI), and data protection-compliant data processing in accordance with the GDPR. The architecture consists of functional modules, each with verifiable, technically measurable results. 4.1 System architecture at a glance

[0015] The platform consists of eight key modules connected via secure APIs, blockchain nodes, and smart contract orchestration. Module 1: Job Aggregation and Semantic Analysis • Function: Extraction and integration of job advertisements from public and private sources. • Method: Natural Language Processing (NLP) for the analysis and classification of job descriptions. • Data sources: LinkedIn, GitHub, StackOverflow, AngelList and internal portals - including but not limited to the sources mentioned. • Classification: Assignment to standardized competency taxonomies. Module 2: GDPR-compliant data processing • Explicit consent: Before any data processing, a granular opt-in consent is obtained in accordance with Article 6(1)(a) GDPR. The consent is digitally signed, timestamped, and stored on-chain as a hash. • Pseudonymization & data masking through AI: Personal data such as name, location, contact details, or gender are automatically pseudonymized before further processing. Identity data is only released after smart contract-verified consent from both parties. 4. Detailed technical description • Self-declared demographic fields (e.g., gender, age, ethnicity) are optional, stored encrypted, and do not contribute to the scoring. The system functions fully even without this information. • Right to deletion via smart contract: Candidates can have their data removed via an on-chain deletion mechanism in accordance with Article 17 GDPR: ◯ Revocation of access tokens ◯ Off-chain confirmation of data deletion ◯ Hash invalidation protocols ◯ Unchangeable audit logs without retention of deleted content • Zero-Knowledge Architecture: Certain credentials (e.g., certificates, skills) are verified by zero-knowledge proofs (ZKPs) without disclosing the raw data, in a privacy-friendly and technically verifiable manner.

[0016] Module 3: AI-based competency assessment • Function: Adaptive, role-based skills tests. • Formats: ◯ Interactive tests ◯ Simulation environments (e.g. programming editors, case scenarios) ◯ Soft skill analysis through behavior-based AI • Storage: Evaluation results are hashed and stored on the blockchain. • ZKP integration: Only the proof (not the result itself) is disclosed. The integration of ZKPs increases the traceability and fairness of the assessment. 4. Detailed technical description

[0017] Module 4: Blockchain-based certificate verification • Structure: Ethereum smart contracts + IPFS off-chain data + Merkle hashing. • Process: ◯ Certificate issuers (e.g. universities, authorities or decentralized networks) sign references digitally. ◯ Qualifications are stored as non-transferable Soulbound Tokens (SBTs). ◯ Verifiers access via public blockchain validations.

[0018] Module 5: AI-based Matching & Ranking • Trust-weighted matching: Candidates are prioritized based on: ◯ Trustworthiness of their references ◯ Job fit probabilities ◯ Bias-corrected AI algorithms • XAI layer: All decisions are provided with explainable logs and stored as a hash on the blockchain.

[0019] Module 6: Smart Contract-based Setting • Escrow payments: Employers must deposit an amount via smart contract before personal data is released. • Contract cycle expiry: 1. Contract terms generated by AI 2. Signature by employer, candidate and moderator (multi-signature model) 3. Release only after confirmed setting 4. Also includes dispute resolution logic and time-limited referral tokens 4. Detailed technical description 4. Supports Ethereum, Polygon and compatible blockchains 5. Architecture-specific circumvention of known patents (e.g. SAP, Workday) 6. Module can be activated or deactivated depending on the jurisdiction.

[0020] Module 7: AI-driven onboarding & career development • Personalized learning paths: Generated from assessment results + target role • Skills updates: Proven new skills are automatically validated and added to the profile. • ZKP-controlled confirmation: Every progress is validated via blockchain and zero-knowledge.

[0021] Module 8: Security, Interoperability and Scalability • Interoperability: API connections to existing HRMS / ATS systems (e.g. SAP SuccessFactors, Oracle HR Cloud). • Security architecture: ◯ End-to-end encryption ◯ Role-based access ◯ On-chain access logs • Scaling: Horizontally scalable microservices with modular deployment • Conformity: Modular design to interact with legally sensitive systems without patent infringement. 5. Patent Classification Codes (IPC)

[0022] The following classifications are used to correctly classify the invention according to the International Patent Classification System (IPC): • G06Q10 / 10 : Administration, business processes, finance or industrial processes: Human resources management, recruitment or scheduling. → Particularly affects job placement, recruitment processes and automation in application management. • G06N20 / 00 : Artificial Intelligence : Machine learning for pattern recognition and classification. → Covers adaptive AI selection processes, bias compensation, competency assessment, and real-time decision making. • G06F21 / 62 : Security measures to protect computers, components or programs : Use of cryptographic techniques for secure communication. → Relevance for blockchain-based certificate validation, identity anonymization, escrow smart contracts, and ZKP-based data protection procedures. • G09B19 / 00 : Education, demonstration : teaching or training procedures. → Captures onboarding automation and AI-driven, role-based training modules upon hire. • G06F40 / 10 : Information retrieval : AI platforms with modular components and reusable functional blocks. → Applies to AI-based systems with integration of adaptive testing, profile generation, and blockchain logging. • G06F16 / 958 : Information retrieval using AI from multiple heterogeneous sources. → Covers systems that capture, aggregate, and analyze job postings, skills data, and credentials from multiple APIs, social platforms, and decentralized networks. 6. Legal compliance and GDPR considerations

[0023] The following mechanisms ensure compliance with European data protection regulations, labor laws, and the EU Artificial Intelligence Regulation through technical default settings (“privacy by design”) and a transparent, auditable architecture: 6.1 Explicit consent & data minimization • Personal data of applicants will be processed exclusively on the basis of voluntary, specific and informed consent in accordance with Article 6 (1) (a) GDPR. • Any data subject may withdraw their consent at any time. Data processing will then be discontinued, and data already processed will be deleted in accordance with applicable erasure requirements. • Personally identifiable information (PII), such as name, gender, location, or contact information, is automatically pseudonymized before being reviewed by employers. Access to PII occurs only after confirmed interest by the employer and separate consent from the applicant. 6.2 Right to erasure (“right to be forgotten”) • Applicants can request the complete deletion of their data - including on-chain references - at any time, as far as this is legally permissible (according to Article 17 GDPR). • The platform uses zero-knowledge proofs (ZKPs) and off-chain methods to protect sensitive data against unauthorized access. • Deletion is carried out in a traceable manner through a user-controlled smart contract call, whereby audit logs are generated without deleted content being stored. 6.3 Algorithmic transparency & auditability • All decisions, evaluations and rankings made by the AI ​​are stored immutably on the blockchain. • The platform uses Explainable AI (XAI) to provide understandable explanations for AI-based decisions - for both applicants and review bodies. • Employers can generate auditable bias audit reports from on-chain logs to meet legal requirements for fairness, transparency, and anti-discrimination. 6. Legal compliance and GDPR considerations6.4 Regulatory compliance

[0024] The platform meets the requirements of: • General Data Protection Regulation (GDPR) of the European Union • General Equal Treatment Acts (AGG / EEO) for freedom from discrimination in recruitment procedures • EU AI Act (2023) for “high-risk AI systems” in the employment context 6.5 Legal and Technical Interoperability

[0025] This system optionally supports integration with the Global Alliance for Workforce AI Governance (GAWA), which the applicant separately filed as a utility model. The GAWA utility model defines a framework for ethical oversight, explainability, and cross-border legal compliance of AI systems. Interoperability between this recruitment platform and GAWA ensures that all hiring processes and decision-making modules are transparent, bias-reduced, and compliant with international AI governance standards. 7. Conclusion & Ethical Considerations

[0026] This invention supports the ethical integration of AI into HR workflows and ensures fair, transparent, and compliant hiring practices. 7.1 Conclusion

[0027] The proposed invention represents a novel and unified system for AI-supported personnel recruitment, qualification verification, and automated onboarding. It addresses long-standing inefficiencies in the recruitment process by integrating the following core functions: • Real-time, AI-supported talent matching, • Verifiable qualification and experience certificates via blockchain-secured certificates, • Completely anonymized, bias-reduced applicant profiles, and • Automated contract processing through smart contracts.

[0028] This system ensures: • Data protection according to GDPR, • Transparency for applicants and employers, • Legally compliant traceability through immutable blockchain protocols, as well as • Scalable, trustworthy personnel development processes.

[0029] The modular architecture allows configurable feature switching to support company-specific legal, operational or ethical requirements. 7.2 Ethical considerations

[0030] The invention follows an “ethics through architecture” approach and ensures compliance with international standards on data protection and fairness: 1. Non-discrimination & bias reduction • Applicant profiles are anonymized in the matching process to avoid biases based on gender, age or origin. • Scoring models are regularly checked for structural or statistical biases using explainable AI (XAI). • Employers can activate diversity-conscious rankings if they wish. 7. Conclusion & Ethical Considerations 2. Gender Equality • Automatic removal of gender-biased data before evaluation. • Candidates can optionally self-identify their gender; this data is encrypted and can only be used for voluntary diversity reports. • Performance evaluations are based on neutral indicators. 3. Privacy-centric AI design • GDPR, EU AI Act and BDSG compliant. • No data processing without explicit consent. • Full access, deletion and export rights for users. 4. Security & Traceability • All decisions, evaluations and contract executions are documented on the blockchain in an audit-proof manner. • No commercial sharing of user data. • No unauthorized profiling outside of the intended purposes. 9. Technical diagrams and process flow descriptions

[0031] To support the claims and to clarify the technical processes, the following diagrams are suggested: Fig. 1: System architecture of the HR platform (overview) Components: 1. AI-powered job aggregation engine 2. Competency assessment module 3. Blockchain layer (Ethereum) 4. GDPR-compliant data storage module 5. AI matching and scoring engine 6. Smart Contract Module 7. AI-powered onboarding assistant

[0032] Process: 8. Job advertisements are aggregated from external sources 9. Candidate submits anonymized profile 10. AI performs real-time assessment 11. Reviews are stored on-chain 12. Employers see anonymized candidate suggestions 13. Escrow payment triggers identity release 14. Smart Contract finalizes the contract 15. Onboarding module is activated 9. Technical diagrams and process descriptionsFigure 2: Candidate selection process

[0033] Steps: 1. Job advertisement is indexed 2. Candidate applies / is nominated 3. AI conducts competency assessment 4. Rating + identity are recorded on blockchain 5. Profile is anonymized and ranked 6. Employer pays escrow → PII is released 7. Smart Contract is initiated 8. Contract is signed, payment is triggered 9. Onboarding content is provided Figure 3: Logic flow of the smart contract trigger: • If employer accepts candidates → Escrow will be blocked • If candidate is confirmed → identity is decrypted • If employment is confirmed → funds will be released • Optional: Refund logic in case of non-employment

[0034] Smart contract fields: • Employer ID • Candidate hash ID • Job ID • Escrow amount • Evidence of qualification verification (hash on blockchain) 9. Technical diagrams and process flow descriptions Figure 4: GDPR-compliant data lifecycle

[0035] Components: 1. Module for obtaining consent 2. Pseudonymization module (using AI) 3. Logging of consent (hash on blockchain) 4. Anonymized AI analysis pipeline 5. Right to erasure trigger with verifiable confirmation of erasure 10.Technical processes & diagram concepts

[0036] The following descriptive steps represent the central technical processes of the system. They serve to visualize and list the patent application as technical drawings. 10.1 System Architecture Flow (Figure 5)Step 1: Job & Candidate Data Aggregation 1. An AI-powered scraping and API engine captures job postings and applicant data from public and private sources. 2. The data is semantically analyzed, classified into taxonomic categories and temporarily stored. Step 2: Consent & Anonymization 1. The platform requests explicit consent via the user interface. 2. After consent, PII is removed through AI-driven pseudonymization before screening or matching occurs. Step 3: AI-based screening & competency assessment 1. Combination of resume analysis, real-time competency tests and simulation tasks. 2. Adaptive assessments tailored to professional roles 3. Results are stored privately off-chain and stored on-chain as a hash for integrity verification. Step 4: Anonymous, AI-supported matching 1. AI evaluates candidates using confidence-weighted scores. 2. Employers only receive anonymized profiles without contact details. 10. Technical Processes & Diagram Concepts Step 5: Smart Contract Hiring Process 1. The employer deposits a security deposit in an escrow smart contract. 2. After mutual consent, the identity is released and an employment contract is automatically concluded. 3. An employment contract is automatically concluded 4. Payment and contract activation are carried out automatically via blockchain. Step 6: AI onboarding & role-specific training 1. AI creates personalized learning paths based on assessment results and job description. 2. Training progress is reviewed and validated and integrated into the profile. 10.2 Application process (Figure 6) 1. Application or sourcing through system 2. Consent request & data processing 3. AI screening + competency test 4. Match evaluation → Shortlisting by employer 5. Escrow payment → identity release 6. Smart Contract is initiated 7. Contract is concluded digitally 8. Onboarding and training are started 10.3 Blockchain verification process for qualifications (Figure 7) 1. An external qualification (e.g. university certificate) is verified off-chain using Zero-Knowledge Proof (ZKP). 2. The corresponding hash is stored on the blockchain. 3. A trusted blockchain oracle component can optionally be used to confirm institutional proofs. 4. Verifying third parties can check authenticity without access to the underlying raw data. Technical drawings

[0037] The following pages contain **technical diagrams, flowcharts, and illustrations** that visually support the systems and processes described in the utility model application. The diagrams are numbered and linked to the corresponding sections in the main document. Fig. 1: System architecture of the HR platform (overview) Process Fig. 1 Fig. 2: Candidate selection process Fig. 3: Logic flow of the smart contract Trigger Smart Contract fields Fig. 4: GDPR-compliant data lifecycle components Fig. 5: System architecture flow Fig. 6: Application process Fig. 7: Blockchain verification process for qualifications

Claims

[1] An AI-powered HR recruitment platform system that aggregates job postings in real time using API and web scraping technologies, classifies roles using NLP, and ranks candidates using trust-weighted AI models based on blockchain-verified qualifications, thereby achieving increased transparency, data protection, and verifiable fairness over traditional HR systems, wherein the components can be used individually or in any combination without departing from the scope of the invention. [2] A GDPR-compliant data management system where candidate data remains anonymised until explicit consent verified by smart contract is given, with revocable access and selective storage on the blockchain, where the components can be used individually or in any combination without departing from the scope of the invention. [3] An AI-driven real-time competency assessment engine that generates role-specific assessments and stores hashed assessment results on a blockchain, enabling privacy-preserving, tamper-proof verification using zero-knowledge proofs (ZKPs), achieving increased transparency, data protection, and verifiable fairness over traditional HR systems, where the components can be used individually or in any combination without departing from the scope of the invention. [4] A decentralized identity system that uses non-transferable Soulbound Tokens (SBTs) to store verifiable professional experience, endorsements, skills, and qualifications, where tokens are issued only after digital confirmation from multiple sources, where the components can be used individually or in any combination without departing from the scope of the invention. [5] An explainable AI (XAI) matching system that stores candidate rankings using immutable on-chain logs and provides audit trails and fairness indicators for employers, where the components can be used individually or in any combination without departing from the scope of the invention. [6] An Ethereum-based multi-signature smart contract model for employment contracts that requires mutual authentication by employer, candidate, and system moderator before payment initiation, where the components can be used individually or in any combination without departing from the scope of the invention. [7] A self-running blockchain-based referral token system tied to smart contracts, whereby referral rewards are paid out only after reaching a predefined period of employment, whereby the components can be used individually or in any combination without departing from the scope of the invention. [8] A candidate profile update protocol that uses AI to initiate changes and secures verification and storage using blockchain and cryptographic signatures, where the components can be used individually or in any combination without departing from the scope of the invention. [9] An AI-based matching algorithm that dynamically adjusts scoring weights based on historical success data and blockchain-verified recommendations, where the components can be used individually or in any combination without departing from the scope of the invention. [10] A privacy-friendly evaluation protocol that stores anonymized interview feedback and assessment results and includes encrypted access layers and traceable audit trails, where the components can be used individually or in any combination without departing from the scope of the invention. [11] An AI-powered onboarding and training module that generates adaptive learning content and tracks skill progress with blockchain-verified milestone validation, where the components can be used individually or in any combination without departing from the scope of the invention. [12] A smart contract enabled dispute resolution system for employment contracts, comprising arbitration logic and evidence histories immutably stored on the blockchain, the components of which can be used individually or in any combination without departing from the scope of the invention. [13] System according to one of the preceding claims, comprising a technical device or software component that automatically masks gender-related candidate information before evaluation or ranking in order to improve DEI (Diversity, Equity, Inclusion) conformity and reduce algorithmic bias in selection. [14] The system of claim 1, further comprising an optional self-identification module allowing candidates to voluntarily indicate their gender, which data is stored cryptographically and used exclusively for anonymised diversity reports or affirmative action, with user-controlled access control.

Images