Mo Yidi (Hazel)

Montreal, Canada

• Built a test dataset of investor KYC documents in Python (pandas, PIL) for an AI tool that reads onboarding documents, combining sample passports and ID cards with blank subscription agreements from 4 issuers and labelling every field to measure accuracy.
• Produced digital, scanned and phone-photo versions of each subscription agreement with planted missing or incorrect fields, testing whether the tool could read poor-quality uploads and flag incomplete files for analyst review.
• Built an Excel KYC requirements checklist from Cayman Islands (CIMA) anti-money-laundering rules, covering required documents and due-diligence level for 11 investor types, used as the rule set for automated missing-document checks.
• Built an Excel risk-rating model that scores investors on PEP status, sanctions, adverse media, industry and country risk, assigns Low/Medium/High ratings, and applies override rules for high-risk cases.

Guangzhou, China

• Built a fund cash-flow schedule from investor reports and SAP records, tracking RMB 139.58M in capital contributions, RMB 136.46M in investments and RMB 146.14M in recoveries to assess capital deployment.
• Reconciled financial statements with Excel VBA macros, comparing distributions with cash flows, investment balances with portfolio assets, and realized gains with income-statement items before reporting.
• Analyzed liquidity and cash-flow timing across contributions, investment payments and recoveries to help the team assess available cash and upcoming funding needs.
• Built Tableau dashboards on capital deployment, cumulative recoveries, distribution progress and remaining portfolio value, used to prepare fund performance summaries for investor updates.

Guangzhou, China

• Reconciled dividend announcements with internal settlement records for eight Hong Kong-listed companies using PivotTables and INDEX-MATCH, verifying payment dates, amounts and posting status.
• Ran monthly variance and trend analysis on dividend data with Power Query, identifying recurring issues and helping reduce monthly exceptions by three cases.
• Investigated delayed dividend payments with SQL, comparing issuer announcements with internal settlement status and documenting open items for operations follow-up.
• Mapped approval and booking steps for dividends, option exercises and name changes against Bloomberg announcements and revised the checklists, increasing peak-period processing capacity by 30%.

Guangzhou, China

• Analyzed monthly loan collections against repayment schedules by borrower and due date, and summarized payment delays and their cash-inflow impact in Power BI dashboards.
• Supported interest-income forecasting by estimating how changes in outstanding balances would affect monthly interest revenue against budget.
• Reviewed 200+ related-party transactions for counterparty identity, amounts, approvals and supporting documents, and summarized exceptions affecting reported credit exposure.
• Ran sensitivity analysis on borrower repayment assumptions using sector and policy trends to assess effects on expected collections and loan-loss provisions.

Guangzhou, China

• Reconciled daily NAV and yield data from iFind and Wind against internal records for 20+ funds, flagging and documenting corrections before client reporting.
• Automated recurring fields in daily reporting schedules with lookup formulas, cutting manual verification by more than 3 hours a day.
• Compared historical returns and NAV trends for 20+ products against peer benchmarks to explain performance differences and support product recommendations.
• Analyzed campaign sales by fund category and client segment and tailored campaign messaging, contributing to a 21% increase in sales.