πΌ Work Experience
Customer Service Advisor, International Markets
The Hut Group (THG) Full Time
β’ Identified gaps in operations reporting and built Power BI dashboards on reviews, refunds, complaints, returns, and claims, used by Customer Service Managers for operational decisions.
β’ Traced recurring complaints to delivery-partner issues and recommended evaluating alternative logistics providers; built Excel reports adopted for escalation tracking across UK, EU, and Southeast Asia.
Finance Manager
ASUSTeK Computer
β’ Automated recurring financial reports with Excel VBA, cutting preparation from 2β3 hours to about 30 minutes.
β’ Built forecasting models checked against prior-year actuals, improving budget accuracy by 15%.
π Education
Master - Computer Science
Westcliff University
2025 - 2027
Master - Financial Technology
University of the West of England
2021 - 2022
π Portfolio & Projects
V.E.T. β Vendor Entity Threat Detection: An AI-Powered Data Fusion Framework for U.S. Supply Chain Security
This paper introduces V.E.T. (Vendor Entity Threat Detection), an AI-powered data fusion framework designed to identify high-risk entities β including shell companies, sanctions-listed organizations, and adversarial front companies β embedded within U.S. commercial and government procurement supply chains. The V.E.T. system addresses a documented gap in existing vendor vetting methodologies, which rely predominantly on manual review, siloed databases, and reactive auditing processes inadequate against sophisticated threat actors employing complex corporate structuring techniques.
V.E.T. proposes a multi-layer machine learning architecture that cross-references six primary data domains: corporate registry records, beneficial ownership disclosures, financial transaction patterns, logistics and shipping data, sanctions list databases (OFAC, BIS Entity List, CAATSA), and open-source intelligence (OSINT) feeds. The target performance threshold is a precision above 85% on high-risk vendor classification, to be validated against a curated benchmark dataset. The framework is aligned with requirements established under Executive Order 13873, NSPM-33, and NIST SP 800-161 Rev. 1, which address ICT supply chain security, federal research security, and supply chain risk management, respectively, in U.S. government and critical infrastructure contexts.
The paper presents the system architecture, data integration design, risk-scoring methodology, and a phased proof-of-concept implementation plan. V.E.T. is built as a shared service accessible through standard API integration, suitable for adoption by federal procurement agencies, financial institutions, and critical infrastructure operators without dependency on any single vendor.
Keywords: supply chain security, AI risk detection, vendor vetting, data fusion, KYC/AML, national security, machine learning, entity resolution, supply chain resilience, critical infrastructure protection, AI governance, beneficial ownership intelligence
A.C.E. β AI Concierge & Engine An AI-Driven Financial Intelligence Framework for U.S. SME Restaurant Resilience
This paper introduces A.C.E. (AI Concierge & Engine), an AI-powered financial intelligence framework designed to address the "Financial Blindness" crisis afflicting U.S. small and medium-sized enterprise (SME) restaurant operators. In the U.S. Accommodation and Food Services sector, 572,464 small employer firms employ more than 8.6 million workers (U.S. Small Business Administration Office of Advocacy, 2025), yet SME restaurant operators work with a structural informational disadvantage relative to large corporate chains: manual invoice processing, fragmented data systems, and reactive pricing decisions create dangerous lags in response to supply chain cost volatility. A.C.E. proposes an Intelligent Neural Network architecture integrating Generative AI and machine learning directly into existing cloud-based point-of-sale (POS) infrastructures β including Toast and Clover β to achieve zero-touch invoice processing via AI-optical character recognition, real-time dynamic menu price optimization, and predictive labor scheduling. The system targets near-elimination of manual invoice data entry (with low-confidence items routed to manual review), a reduction of 15+ administrative hours per location per week, and measurable improvement in pricing decision velocity during inflationary periods. This paper presents the problem diagnosis, system architecture, AI component design, implementation strategy, risk assessment, and ethical governance framework. A.C.E. connects to the cloud-based POS systems restaurants already use, so independent operators can adopt it without replacing hardware.
Keywords: AI Restaurant Technology, SME Financial Resilience, Invoice Automation, Dynamic Pricing, Predictive Labor Scheduling, Generative AI, POS Integration, National Economic Impact, Financial Blindness, AI Governance
π Certifications
- Google Advanced Data Analytics Professional Certificate - Google