Skip to content
View Romequinco's full-sized avatar

Block or report Romequinco

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Content in all repositories owned by your account will be closed.
Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. Markdown is supported. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
Romequinco/README.md

Hi, I'm Óscar

Quantitative analyst · Financial markets & AI · Bilbao, Madrid or remote

Website LinkedIn

I spent three years executing equities on a trading desk. Now I do quantitative research and build AI tools for markets: they explain what happened, check their own figures and never tell anyone what to buy.

Auguria dashboard
Auguria
Market analysis in Spanish · in production
Briefly app
Briefly
A two-voice podcast on your stocks · live demo
STATOR dashboard
STATOR
Liquidity engine for illiquid stocks · paper trading

Selected work

  • Auguria: Market analysis in Spanish. 1,071 commits in five months.
  • Briefly: A two-voice podcast on your stocks. 232/232 figures traced to a source.
  • STATOR: Liquidity engine for illiquid stocks. 0 of 2,938 orders out of limits.
  • 10-K agent: Answers questions from SEC filings. 8/10 blind questions right.
  • Market regimes: 12 detectors, 60 years of data. 16/17 crises caught.
  • Document systems for companies: Ordered archives and an AI assistant for 5 companies. In use.

All 30 projects, each with figures and a write-up, are on romequinco.github.io.

Stack

Python TypeScript PyTorch scikit-learn pandas LangGraph FastAPI Next.js Supabase Docker

En español

Analista cuantitativo de mercados financieros. Tres años ejecutando renta variable en una mesa; ahora hago investigación y herramientas de IA para los mercados. Fundador de Triarii Consulting y alumno del máster MIAX (Instituto BME). Todo mi trabajo, con figuras y explicación: romequinco.github.io/es

Pinned Loading

  1. PRUEBAS_DETECCION_REGIMENES_DE_MERCADO_TFM PRUEBAS_DETECCION_REGIMENES_DE_MERCADO_TFM Public

    Leak-free, walk-forward benchmark of 12 market-regime detectors over 1962–2026. Simple rules won: 16 of 17 crises caught. Part 1 of my MIAX master's thesis.

    Jupyter Notebook

  2. fair-credit-risk-nn fair-credit-risk-nn Public

    Neural credit-risk model that cuts the gender gap by 72% for 0.6 AUC points, with calibrated uncertainty (MC Dropout) on Home Credit data. MIAX, team of 3.

    Jupyter Notebook

  3. CHATBOT_ORGANIZACION CHATBOT_ORGANIZACION Public

    Telegram agent that turns team chat and voice notes into tracked tasks, with daily summaries at 8:30 and 17:00. In production at Albujón Energy.

    Python

  4. 10k-financial-agent 10k-financial-agent Public

    LLM agent that answers questions on SEC 10-K filings, cites the exact source and checks every figure against XBRL. 8/10 blind questions right. MIAX, team of 3.

  5. equity-arbitrage-oportunities equity-arbitrage-oportunities Public

    Consolidated order book across BME, Aquis, Cboe and Turquoise to find cross-venue arbitrage: 2,719 opportunities, wiped out by a few milliseconds of latency. MIAX.

    Jupyter Notebook

  6. multimodal-market-briefer multimodal-market-briefer Public

    Briefly: your stocks' close as a four-minute, two-voice podcast. Reads PDFs, charts and broker screenshots and answers by voice. 232/232 figures traced to a source. MIAX, team of 3.

    Python