PhD in Mathematical Finance & Stochastic Calculus, MS in Software Engineering—architecting ML platforms and Generative AI solutions at enterprise scale for banking and tech.
I'm a Data Scientist Manager at Banamex with a unique blend of deep mathematical expertise and cutting-edge AI engineering skills. My PhD in Mathematical Finance & Stochastic Calculus, combined with a Master's in Software Engineering, enables me to bridge the gap between theoretical innovation and practical implementation.
Currently, I lead the development of an enterprise ML platform similar to Uber's Michelangelo, integrating Generative AI solutions that process 1,000+ daily customer interactions with 95% accuracy. My work has reduced time-to-production by 60% and operational costs by 30%.
My research in quantitative finance and mathematical modeling has been published in peer-reviewed journals, and I continue to apply advanced stochastic calculus and machine learning techniques to solve complex real-world problems.
Python, R, JavaScript, TypeScript, SQL
TensorFlow, PyTorch, Scikit-learn, MLflow
OpenAI GPT-4, Whisper, Google Gemini
PySpark, Hadoop, Cloudera
FastAPI, Flask, PostgreSQL
AWS, Heroku, Airflow, Docker
Universidad Nacional Autónoma de México
Focus: Mathematical Finance, Machine Learning, Probability & Statistics
Advisor: Prof. Pablo Padilla Longoria
UNAM-BBVA Foundation AwardUniversidad de los Andes
Focus: Algorithm Design, Software Architecture, Full-Stack Development, DevOps
University of Texas at Austin & GreatLearning
Universidad Nacional Autónoma de México
Focus: Probability Theory, Statistics, Mathematical Finance
Advisor: Prof. Ramsés H. Mena
Pontificia Universidad Javeriana
Minor: Economics | Focus: Mathematical Analysis, Econometrics
Advisor: Prof. Gerardo Román Chacón
Independent Contractor
Remote
Developed an MVP for a Praktika-like language learning application featuring real-time conversational AI for grammar practice. Built microservices architecture using Docker containerization, FastAPI for API endpoints, and VLLM as LLM server provider with advanced prompt engineering and guardrails implementation. Created robust backend infrastructure with Django and PostgreSQL for user management, course creation, and chat history retrieval. Designed responsive frontend using React, Vite, and Tailwind CSS for seamless user experience.
Architected and deployed an enterprise-grade RAG (Retrieval-Augmented Generation) document Q&A chatbot for medical literature, similar to OpenEvidence. Implemented backend with FastAPI and PostgreSQL with pgvector extension for semantic search capabilities. Orchestrated RAG pipeline using LangChain with AWS Bedrock (Llama 3-70B model and Amazon Titan embeddings) for accurate medical query responses. Built automated PDF ingestion pipeline from S3 bucket processing medical documents into vector embeddings. Developed React 18 + TypeScript frontend with Vite, implementing chat interface and conversation history management. Containerized entire stack using Docker Compose with hot-reloading for development efficiency.
Banamex
Mexico City, Mexico
Leading the design and development of an enterprise ML platform similar to Uber's Michelangelo, integrating FastAPI, MLflow, Spark, TensorFlow, and Scikit-learn to accelerate model development and deployment processes across the bank.
Developed ETL pipelines integrated with ML/NN models using PySpark, TensorFlow, and Scikit-Learn for advanced feature engineering, improving credit model KPIs by up to 5%.
Designed and deployed the first enterprise recommender system using ALS in PySpark, integrated with Adobe Ecosystem and AWS services, solving cross-border data transfer challenges while serving personalized content to mobile applications.
Designed and implemented Airflow orchestration on Cloudera for automated capacity monitoring, triggering alerts when databases reach maximum capacity on Hadoop clusters and RHEL servers.
Built resilient FastAPI applications for ETL pipeline orchestration, enabling efficient resource management and automated scaling.
Developed an ML-powered ATM location optimization application using FastAPI and PyArrow for efficient data lake access, featuring interactive Google Maps-like visualization and WhisperAI voice search capabilities.
Universidad Nacional Autónoma de México
Mexico City, Mexico
Conducted cutting-edge research in quantitative finance, applying machine learning techniques and stochastic calculus to develop novel algorithmic trading models, resulting in peer-reviewed publications in top-tier financial journals.
Co-designed and implemented sophisticated pricing algorithms for complex financial derivatives in American markets using Python and R, achieving improvement in pricing accuracy compared to traditional Black-Scholes models.
Applied advanced stochastic differential equations and Monte Carlo simulations to model market volatility and risk assessment, contributing to award-winning doctoral thesis recognized by UNAM-BBVA Foundation.
Authors: Ivan D. Peñaloza-Rojas, Pablo Padilla Longoria
Computational Economics, Springer, Netherlands, 2021
Developed novel pricing algorithms for financial derivatives in markets with incomplete information and constraints, combining stochastic calculus with machine learning techniques.
Type: PhD Dissertation, 2021
Universidad Nacional Autónoma de México
UNAM-BBVA Foundation AwardType: Master's Thesis, 2016
Universidad Nacional Autónoma de México
Type: Bachelor's Thesis, 2012
Pontificia Universidad Javeriana
I'm always interested in discussing new opportunities, collaborations, or innovative projects in ML/AI and data science.