Quantitative Systems Engineer | Full-Stack Architect | Algorithmic Trader
I build production systems where statistical edge meets code excellence.
5+ years of experience in low-latency trading infrastructure, MLOps, and distributed systems.
Tickflow Capital's proprietary in-house platform for systematic trading.
- Engineered an event-driven backtesting engine processing 7 years of 1-minute tick data, achieving a 75% walk-forward pass rate (4 out-of-sample windows, 2018–2025) with a Sharpe ratio of 2.10 and avg profit factor of 1.90 across 124 trades.
- Architected a 7-layer risk validation gate (kill-switch, drawdown limits, slippage monitoring, preflight validation), enforcing hard daily budget limits and reducing anomalous execution risk to 0% in live paper-trading.
- Integrated MetaTrader 5 for sub-second order dispatch with exponential backoff retry logic and fill reconciliation.
- Deployed a full observability mesh using Prometheus/Grafana/Loki (12 metrics, 9 alert rules) paired with an immutable, append-only PostgreSQL audit system enabling full trade-decision replay via
trace_idcorrelation. - Containerized with Docker and orchestrated on AWS, enabling horizontal scaling for Monte Carlo ensembles of 10,000+ iterations.
Validation Highlights:
- Best Window Return: +13.14% (Window 1, 2020)
- Monte Carlo Confidence: 98.6th percentile
Open-source (MIT) production platform for stock direction prediction.
- Developed a hybrid forecasting model (50% LSTM / 30% FinBERT sentiment / 20% technicals) serving 10 symbols with 15+ years of historical data.
- Designed a REST API gateway with 30+ endpoints (OpenAPI/Swagger) using dual-layer authentication (JWT + PBKDF2-hashed keys) and tiered rate-limiting.
- Orchestrated async ETL pipelines (Celery + Redis) for multi-source news ingestion, applying SHA-256 deduplication to reduce redundant processing by an estimated 40%.
- Maintained >99% service availability for automated 7-day prediction tracking via Sentry + UptimeRobot.
| Domain | Technologies |
|---|---|
| Quant & ML | Python, PyTorch, scikit-learn, Pandas, NumPy, FinBERT, MetaTrader5 |
| Backend | Django, FastAPI, Celery, Redis, PostgreSQL, WebSockets |
| Frontend | React, Chart.js, TailwindCSS |
| Infra & DevOps | Docker, AWS (EC2/RDS), Cloudflare, GitHub Actions, Nginx |
| Observability | Prometheus, Grafana, Loki, Sentry |
I enjoy deep dives into walk-forward validation, risk architecture, async Python, and open-source tooling.
Building systems that don't lie.