Kharkiv Metropolitan Area
Most market data is noise. I build systems that filter IPO and insider activity into actionable signals. • IPO dataset (~3,400 IPOs, 40y) • SEC filings parsing (S-1, 424B4) • Return reconstruction (1d / 30d / 90d) • Real-time insider signal pipeline: – Form 4 ingestion and normalization – trade scoring (size, anomaly, insider behavior) – market context enrichment Signals are used for: • discretionary trading • algorithmic strategies • backtesting and research Goal: turn raw financial data into reproducible market signals. Focus: IPOs, insider trading, market signals, data-driven investing, quantitative research
Building data-driven systems to analyze IPOs and insider trading activity. • Developed IPO dataset (~3,400 IPOs, 40y) • Parsing SEC filings (S-1, 424B4) • Reconstructing returns (1d / 30d / 90d) • Built real-time insider signal pipeline: – Form 4 ingestion and normalization – Trade scoring (size, anomaly, insider behavior) – Market context enrichment • Signal output: – Telegram (filtered ideas) – Redis (for automated strategies) Goal: transform raw market data into reproducible trading signals.
Building a data-driven signal system based on insider trading (Form 4), congressional trades, and IPO data. Processing millions of events → filtering into a small set of high-quality signals. Focus: – identifying abnormal insider behavior – filtering out noise (~80%+ of raw events removed) – tracking signals in real time + follow-ups Key insight: Raw data ≠ edge Edge = filtering + context + timing Currently testing: – real-time signal delivery – execution vs theoretical performance – IPO + insider combined frameworks Goal: Turn public market data into actionable, structured signals.