Portfolio
Research and tools across Korean real estate, forensic finance, and procurement markets. Each project documents the problem it solves, the data sources it uses, and the decisions behind the design.
17 items · 5 featured
Korean Forensic Finance
Featured
Ongoing
Korean Forensic Accounting Toolkit A 13-repo, 749-test forensic accounting platform that screens Korean DART filings for earnings manipulation and convertible bond dilution — from raw government data to investigative-grade output, automated end to end. Each component is independently published, MIT-licensed, and accompanied by a stand-alone write-up. View case study Ongoing
KOSDAQ CB/BW Dilution Screen 2,934 KOSDAQ CB/BW issuances scored across four calibration runs; flag rate reduced from 49.3% to 33.1% after resolving a systematic denomination mismatch between split-adjusted prices and DART exercise prices. Four cases confirmed as genuine extreme ITM at above 10x moneyness. View case study Ongoing
Korean Beneish M-Score (kr-beneish) Korean IFRS adaptation of the 8-variable Beneish M-Score calibrated against a 50-case labeled dataset (17 FSS/SFC fraud, 13 clean, 20 auto-controls); Korean threshold -2.45 vs. US -1.78; flags ~1,250 of 7,447 KOSDAQ company-years (2018–2023). View case study
Featured
Korean Accounting Enforcement Dataset 240 Korean accounting violations coded across FSS and SFC enforcement records, DART-linked and validated through a five-phase bias audit — the first open, reproducible Korean enforcement dataset with machine-readable violation taxonomy and Beneish ratio coverage for named companies. View case study Ongoing
krff-shell — Natural Language Finance Shell 11-tool MCP server exposing the Korean forensic finance toolkit as natural-language queries via Claude Desktop — DuckDB query layer, parameterized SQL (injection-safe), per-company self-contained HTML reports (Plotly), optional Claude narrative synthesis. FastAPI REST endpoints mirror all 11 MCP tools. 317 tests. View case study Korean Real Estate
Featured
Ongoing
Korean District Credit Risk Index A 16-year, 228-district time-series of Korean housing credit risk built from public court registry data — tracking the five registration types that lead and lag residential foreclosure, with a statistically consistent 17-month cascade from lease-lien onset to forced auction. View case study Ongoing
Korean Apartment Transaction Anomaly Screen A statistical screen using only public MOLIT data to flag apartment buildings where coordinated transaction cancellations suggest price manipulation. Demo: 강남구 2024 — 3,754 transactions analyzed, 33 buildings flagged. Sector findings cross-validated against public FSS and 국토부 investigation records. View case study Ongoing
Korean Housing Risk — Public Dataset The public storefront for the Korean District Credit Risk Index — sample district reports, a national 16-year trend chart, and methodology documentation, published under CC BY-NC 4.0. The first publicly available 시군구-level court registry time series covering 228 Korean districts. View case study Ongoing
Korean Presale Transfer Window Scanner A data pipeline that crosses 청약홈 presale lottery records with K-APT construction permit data to surface apartments where 분양권 전매제한 (presale transfer restrictions) have expired — a transaction window that 청약홈 never surfaces directly. 6 APIs confirmed live. 10/10 join test passed. View case study Ongoing
Korean Lease Risk Engine V0 CLI proof-of-concept: resolves a Korean rental address to a district-level distress band using 16 years of court registry data, positioning the district on the 17-month 임차권→강제경매 cascade timeline. Not calibrated for paying-tenant decisions — V0.5 required for property-level analysis. View case study Ongoing
Korean Public Auction Intelligence Market research and product specification for a Korean public auction scoring tool targeting KAMCO's 온비드 platform. 86,875 unsold listings in 2024 — 2× the 2023 count. Primary signal: 유찰횟수 (failed auction rounds), which depresses reserve prices in a structured, predictable pattern. All four scoring fields confirmed in the KAMCO API. One API test away from build decision. View case study Ongoing
Korean Real Estate Tax Screen Screens Korean apartment transaction records for cancellation-cluster anomalies associated with price manipulation and tax evasion, using the MOLIT 실거래가 public API. Module A is live with 37 passing tests. View case study Ongoing
Korean Tenant Transaction Assurance — Policy Research A policy proposal for 임차인단독의뢰형 공공거래확인제도 — a public registry verification service allowing tenants to independently order official property checks before signing a lease, without landlord cooperation. Submitted to the 한국부동산원 Housing-Rights Idea Competition (2026-05-08). The proposal directly motivated the technical development of lease-risk-engine as a working proof-of-concept. View case study Ongoing
Korean Appraisal Tech — Market Research Systematic market research into software opportunities in the Korean real estate appraisal sector. Four hypotheses evaluated across six research rounds using an invalidation-first method. Hypothesis A produced a working demo — now the independent fraud-screen project. Hypothesis B (solo 감정평가사 workflow tool) is research-more, pending one practitioner interview. View case study