A personal archive of work in mathematics, markets, systems, education and technology. Each record opens with what was built, then the question that drove it.
Book a callSummarize https://ionitsa.com/ for me. Describe its purpose, main contents, and evidence of technical work. Separate statements made by the site from your own assessment, note anything unclear or unverified, and avoid promotional language. Optional content index: https://ionitsa.com/llms.txt
Reproducible research into cash-versus-synthetic credit dislocations, beginning with one issuer and expanding into stress events and CDX rolls.
Can issuer-matched cash bonds and CDS isolate a repeatable relative-value return after rates, financing, liquidity, and event risk?
Domains Hedge Fund, Credit Methods Hedging, Risk Modeling, Backtesting, Event Studies
A tested CDS hazard-curve and fixed-rate bond valuation library that turns credit quotes into survival probabilities, full-revaluation P&L, and model-validation evidence.
When do linear spread-duration approximations cease to be reliable for CDS valuation, bond P&L, and bond–CDS hedge sizing?
Domains Hedge Fund, Credit Methods Derivatives Pricing, Curve Bootstrap, Numerical Methods, Stress Testing
A production liquidity score that combines Markit and Bloomberg market evidence with position concentration, liquidation capacity and point-in-time stress conditions.
How can vendor liquidity signals, fund position sizes and point-in-time market stress be combined into one comparable and explainable measure of executable credit liquidity?
Domains Hedge Fund, Credit Methods Risk Modeling, Data Fusion, Stress Testing, Portfolio Analytics
A reproducible US/EU pilot combining actual daily equity and credit-factor data, current ICE CDS settlements, event studies, implementation-aware hedge tests, and Monte Carlo diagnostics.
Can broad, issuer-level equity stress improve the timing of credit hedges, and can the relationship ultimately be traded through bonds or CDS?
Domains Hedge Fund, Credit, Equities Methods Entity Resolution, Event Studies, Bootstrap, Backtesting
Latency-sensitive earnings-release intelligence for ~2,000 bond issuers—forecasting disclosure windows, monitoring issuer pages, and capturing PDFs before standard feeds.
How accurately can disclosure timing be forecast from historical release behaviour and issuer-specific reporting patterns?
Domains Hedge Fund, Credit Methods Forecasting, Event Risk, Monitoring, Workflow
Hedge fund compliance system for labeling bond execution quality across Bloomberg, Markit, dealer quote, and trade-print benchmarks.
How can execution quality be classified fairly when bond trades must be compared against several incomplete market benchmarks?
Domains Hedge Fund, Credit, Compliance Methods Scoring, Execution Analysis, Controls, Market Data Normalization, Benchmark Selection
Hedge fund trade-capture parser that converts semi-structured Bloomberg IB messages into normalized trade events across multiple asset classes.
How can semi-structured human trade messages be converted into deterministic events without losing ambiguity or provenance?
Domains Hedge Fund, Execution, Post Trade Methods Regex Extraction, Classification, Normalization, Trade Capture
Hedge fund execution pipeline that normalizes FIX messages into validated market events for analytics and controls.
How can stateful FIX messages be normalized into reliable market events while preserving ordering, validation, and traceability?
Domains Hedge Fund, Execution Methods Data Ingestion, Data Quality, Normalization
PDF debt-note extraction and Markit bond matching—linking ISIN-level disclosures in filings to issuer reference data and quarterly financials.
How can unstructured debt disclosures be linked to canonical bond records with an auditable level of confidence?
Domains Hedge Fund, Credit, Documents Methods Nlp, Regex Extraction, Entity Matching, Data Ingestion
Research-style reconciliation framework that uses trade logs to infer instrument-ID mappings across disconnected financial systems.
How can instrument identities be inferred across disconnected systems when no complete common identifier exists?
Domains Hedge Fund, Post Trade, Portfolio Methods Record Linkage, Fuzzy Matching, Entity Resolution, Controls
PDF extraction pipeline for hierarchical geography and segmental disclosures in financial statements—detecting multi-level table headers and normalizing nested breakdowns into auditable structured feeds.
How can hierarchical financial disclosures be reconstructed reliably when their structure is encoded visually across irregular tables?
Domains Documents, Research Operations, Hedge Fund Methods Pdf Parsing, Table Extraction, Hierarchical Headers, Data Normalization, Auditability
Operational dashboard and lineage layer for multi-model reinsurance runs on Tyche—tracking outputs, runtime, treaty KPIs, and parameter sensitivity.
How can model lineage, runtime behaviour, and parameter sensitivity reveal whether complex reinsurance runs remain trustworthy?
Domains Reinsurance, Insurance Methods Workflow, Observability, Data Ingestion, Monte Carlo, Visualization
Explored anomaly and supervised models on a severely imbalanced card-transaction dataset to rank transactions for analyst review.
Which combination of anomaly detection and supervised learning best ranks rare fraudulent transactions for constrained human review?
Domains Fraud, Payments Methods Anomaly Detection, Feature Engineering, Imbalanced Classification, Isolation Forest
Crypto fraud detection layer combining login/signup behaviour, allocation patterns, and on-chain wallet evidence.
How can weak behavioural and on-chain signals be combined to distinguish coordinated identities from legitimate applicants?
Domains Crypto, Fraud, Identity Methods Graph Analysis, Risk Scoring, Heuristics
Phase-1 audit and scoring prototype for a crypto social platform: heavy-tail distribution analysis, rarity-driven achievement automation, and no-downgrade rollout simulation across 24k+ users.
How should reputation be distributed and updated when user activity follows a heavy-tailed rather than uniform distribution?
Domains Crypto, Identity Methods Distribution Analysis, Rarity Scoring, Feature Engineering, Graph Analysis, Simulation
Post-IPO / ICO investor conviction score measuring token retention after distribution, including wallet balances, staking deposits, and LP exposure.
How can investor conviction be estimated from post-distribution token behaviour across wallets, staking, and liquidity positions?
Domains Crypto, Token Distribution, Investor Analytics Methods Scoring, Behavior Analysis, Onchain Analytics, Time Series
Interpretable NFT marketplace screener combining trait rarity, listing prices, trait floors, and liquidity confidence.
How much explanatory value do trait rarity and liquidity contribute to the relative pricing of unique digital assets?
Domains Crypto, Research Operations Methods Rarity Scoring, Marketplace Analytics, Trait Analysis, Price Screening
Full-stack prototype that scores public X accounts on a transparent 0–100 scale from official API data, with categorized evidence and a shareable report UI.
How can public account quality be measured transparently without collapsing heterogeneous evidence into an opaque score?
Domains Identity, Hiring, Research Operations Methods Risk Scoring, Heuristics, Feature Engineering, Data Ingestion, Observability
Adaptive AI training platform that models a student's knowledge over time, selects the next useful problem, and evaluates written reasoning—not just final answers.
How can a learning system infer a student's changing knowledge state and select the next problem that maximizes useful progress?
Domains Education, Olympiad Prep Methods Adaptive Learning, Knowledge Modelling, Written Feedback
AI navigator for UK and international academic competitions—cataloguing opportunities, tracking deadlines, and recommending next challenges from a verified catalogue.
How can fragmented competition information be transformed into a reliable recommendation path for each student's interests and level?
Domains Education, Competition Planning Methods Recommendation Engine, Catalogue Structuring, Deadline Tracking
Hard-skill extraction from CVs, skill-relationship graph, and profession inference via balanced vector matching.
How can a person's likely profession be inferred from the structure and balance of their observed hard skills?
Domains Hiring, Quant Recruiting Methods Vector Matching, Graph Matching, Scoring, Tf Idf
Interpretable candidate ranking that combines hard-skill fit, environment fit, experience, and constraints—no neural black box.
How can candidate fit be ranked across skills, experience, environment, and constraints while keeping every recommendation explainable?
Domains Hiring, Quant Recruiting Methods Ranking, Scoring, Vector Matching, Tf Idf
Observable behavioural signals from professional platforms used to estimate candidate–environment compatibility over time.
Can longitudinal professional behaviour estimate compatibility with a working environment more reliably than static self-reporting?
Domains Hiring, Quant Recruiting Methods Vector Matching, Time Series, Scoring
Spatial systems optimizer for overlapping entity nodes, using planar ellipse layout and topology constraints.
How can overlapping spatial entities be arranged to preserve topology while minimizing visual collision and distortion?
Domains Geodata, Systems Design Methods Geometry, Optimization, Visualization