AutumnTHE OLYMPIAD COLLECTION
Menu
RESEARCHED WITH AUTUMN
MR

Mateo Restrepo

Senior ML Engineer
N-iX · Medellín, Antioquia, Colombia · engineer

Currently working at N-ix.com as an ML-Engineer on computer vision algorithms (clustering, recomendations, background removal, superresolution) for a big US client. Until Jan 2024, I worked at La Haus, a tech-based residential property market-place in Colombia. Together with my team, we carried out several machine learning / AI related initiatives: such as, a GPT chatbot agent using ChatGPT + RAG, automatic image quality scoring, for ranking of property images in search results, a personalized recommendation engine, to help users find best house for them according to their search parameters and an automated valuation model. I worked for Bancolombia for three and half years as part of the new "Analytic Capabilities Team" (Dirección de Capacidades Analíticas), in the adoption of machine learning based solutions for several of the bank's most pressing problems. Together, we develop analytic models, both supervised and unsupervised, that allow us to extract actionable insights from

Details
LocationMedellín, Antioquia, Colombia
Company siten-ix.com
UniversityCornell University
Machine LearningComputer VisionRecommender SystemsData AnalyticsApplied Mathematics
Notes
  • Quant employment verification: CONFIRMED. Primary-source confirmation: the Cornell ORIE page for advisor Huseyin Topaloglu lists Mateo Restrepo's PhD (Computational Methods for Static Allocation and Real-Time Redeployment of Ambulances, co-advised with Shane Henderson, 2008) with 'First position at Merrill Lynch, New York, NY.' Merrill Lynch (Bank of America Merrill Lynch after the Jan 2009 Bank of America acquisition) is a global investment bank; the index-enrichment role descriptions place him on an Automated Market Making (Listed Equity Options) desk and an earnings-event volatility modeling team, both quantitative finance. Identity is supported: Cornell PhD 2008 + Colombia olympiad background + Medellin/Bancolombia continuity all match one person. The specific titles/dates ('VP Quantitative developer 2008-10 to 2011-10', 'Associate') are index-only and not independently primary-confirmed.2008
  • PhD in Applied Mathematics, Cornell University, 2008; dissertation 'Computational Methods for Static Allocation and Real-Time Redeployment of Ambulances', co-advised with Shane Henderson (Operations Research and Industrial Engineering).2008
  • IMO 1997: representing Colombia; individual result page on imo-official.org (contestant 4773) and moresults.org show a bronze medal, individual rank ~155/460.Jul 1997
  • Mathematics Genealogy Project record for Mateo Restrepo Mejia: Ph.D. Cornell University 2008.2008
  • Career trajectory: quantitative finance (Merrill Lynch/Bank of America, New York, ~2008-2013, incl. Benchmark Solutions) -> mathematical modeling consulting (Math Decision, Medellin, 2013-2014) -> bank analytics leadership (Grupo Bancolombia, 2014-2018) -> head of data/ML (Yuxi Global 2018-2019; La Haus 2019-2024) -> ML engineering (N-iX 2024-present). A durable applied-mathematics thread runs from the Cornell OR PhD through pricing/optimization to ML.
  • DBLP author record (Mateo Restrepo) lists publications including 'A simple GAP-canceling algorithm for the generalized maximum flow problem' (Math. Program. 118(1), 2009, with David P. Williamson) - operations-research / optimization research from the Cornell period.
  • LinkedIn (index enrichment) lists 'Olimpiadas Colombianas de Matematicas - Gold Medallist, Mathematics, 1993-1998' and Colegio Aleman (1984-1997).
  • Enrolled in a Master's in Artificial Intelligence at Universidad de los Andes (2024-2025) while working as an ML engineer, indicating continued formal study alongside industry work.
  • Notes are drawn from a mix of index enrichment and primary sources; index-only claims (e.g. exact BofA/Merrill titles and dates, Benchmark Solutions) are labeled as such and are not independent primary-source confirmation.

Competition record

Colombia · IMO

Related collections

Explore the collection