Machine Learning Engineer · Deep Learning · Applied ML Research
Andres
Cruz Chipol
Building machine learning systems from data and model development to training, evaluation, deployment, and production inference.
MSc in Computer Science
About
Machine Learning Engineer and Computer Scientist focused on Machine Learning, Deep Learning, and ML Systems. My experience spans the entire ML lifecycle — from exploratory data analysis, preprocessing, and feature engineering to model training, evaluation, optimization, deployment, and production inference.
I develop and deploy models across classical ML (gradient boosting, random forests), unsupervised learning (clustering, dimensionality reduction), and deep learning (speech technologies, ASR/TTS), working with Python, PyTorch, TensorFlow, Keras, Scikit-learn, and Hugging Face. I build reliable pipelines and inference services using MLflow, Weights & Biases, Docker, FastAPI, and Linux, with hands-on experience in high-throughput model serving via vLLM and Ollama, as well as LLM-powered applications and agents.
Currently pursuing an MSc in Computer Science at Cinvestav with a focus on Artificial Intelligence, my work centers on Applied ML Research, Deep Learning, and ML Systems engineering — bridging research methodologies with reliable, scalable, and efficient production environments.
Work Experience
Artificial Intelligence Analyst
- Developed and evaluated classical machine learning models using algorithms including Random Forest and XGBoost for real-world applications.
- Performed exploratory data analysis (EDA), feature analysis, preprocessing, visualization, and data preparation for machine learning workflows.
- Applied dimensionality reduction techniques, including PCA, t-SNE, and UMAP, for feature analysis, visualization, and representation exploration.
- Applied unsupervised learning and clustering techniques, including DBSCAN, HDBSCAN, and hierarchical clustering, for customer segmentation and behavioral analysis.
- Built and maintained MLOps pipelines for model training, experiment tracking, evaluation, and deployment using MLflow, Weights & Biases, Docker, FastAPI, and cloud infrastructure.
- Trained and fine-tuned deep learning and speech models using PyTorch and Hugging Face, including ASR and TTS models such as Whisper, WhisperX, PiperTTS, CosyVoice, and Bark.
- Evaluated and optimized trained models for deployment and inference in production-oriented environments.
- Deployed and served LLMs and ML models on Linux-based servers using vLLM and Ollama, supporting staging and production environments.
- Developed ML inference services using Python, FastAPI, REST APIs, and WebSockets for real-time applications.
- Developed LLM-powered applications and intelligent agents using LangChain and LangGraph.
Data Scientist
- Developed and evaluated classical machine learning and deep learning models for early detection and patient referral in oncology.
- Built ML workflows for data preprocessing, model development, evaluation, and inference using Python, TensorFlow, Keras, Scikit-learn, Pandas, and NumPy.
- Optimized the ML inference pipeline, reducing processing times from minutes to seconds by improving inefficient Python-based data processing and execution workflows.
- Integrated machine learning models into backend services using JavaScript, Node.js, and Python.
- Contributed to R&D projects applying machine learning and deep learning to healthcare, translating computational approaches into clinical-oriented applications.
Machine Learning Research Fellow
- Built data collection and preprocessing pipelines for machine learning research projects.
- Developed web scraping solutions in PHP to collect datasets from multiple sources.
- Performed data cleaning, manipulation, and labeling using R and RStudio.
- Applied machine learning algorithms with Python, Pandas, and Scikit-learn for classification and structured-data analysis.
- Developed ETL workflows integrating data extraction, preprocessing, and ML-based classification.
- Contributed to research-oriented machine learning projects, applying computational and data-driven methods to real-world datasets.
Data Science Summer School Participant
- Collected and processed datasets using the Twitter API, developing data extraction and preprocessing workflows.
- Performed data cleaning, transformation, and exploratory analysis using R and the Tidyverse ecosystem.
- Developed sentiment analysis workflows for cryptocurrency-related social media data.
- Applied statistical and machine learning techniques to analyze social media data and extract research-oriented insights.
- Worked with ETL processes and data pipelines for data-driven research projects.
- Received research training in data science, machine learning, exploratory data analysis, and information retrieval.
Featured ML Projects
Applied machine learning research, production inference systems, and deep learning engineering.
View full catalog →
Medical Imaging & Deep Learning VinDr-Mammo: Deep Learning & Vision-Language Modeling for Mammography
Applied deep learning research and experimentation on high-resolution mammography imaging: dataset preprocessing, feature extraction, and vision-language model fine-tuning for BI-RADS assessment.
ML Systems & MLOps Production ML Lifecycle & Multi-Dataset Inference Service
An end-to-end machine learning system covering dataset preprocessing, model training, cross-validation evaluation, model serialization, containerization with Docker, and real-time inference with FastAPI.
Deep Learning & Speech ML Deep Learning Speech Processing & Audio Intelligence
Fine-tuning, evaluation, and optimization of deep learning architectures for speech recognition (ASR) and synthesis (TTS) using PyTorch, Hugging Face, Whisper, and state-of-the-art voice models.
ML Systems & Model Serving High-Throughput Model Serving & GPU Inference Infrastructure
Production-oriented model serving and deployment on Linux systems using vLLM and Ollama. Engineered high-concurrency inference pipelines, WebSocket endpoints, and containerized serving infrastructure.
Stack
From model training to production systems.
Python
PyTorch
TensorFlow
Keras
Scikit-Learn
NumPy
Pandas
Matplotlib
LangChain
LangGraph
MLflow
FastAPI
JavaScript
C++
PostgreSQL
MySQL
Docker
Git
GitHub
Jupyter
VS Code
Bitbucket
SeabornWriting
Technical writing on Machine Learning, Deep Learning, ML Systems, and software infrastructure.
View all →GitHub & Open Source
A year of code, commits, and open-source machine learning repositories.
Featured Repositories
treedecision_e2e_multidataset
Production-grade ML monorepo deploying three classification domains via unified FastAPI, Docker Compose, and automated pytest suites.
reschip/Qwen3.5-VL-BIRADS
Fine-tuned vision-language model for digital mammography BI-RADS assessment with Unsloth and GGUF quantization (700+ downloads on HF).
Education
Master of Science in Computer Science
Artificial Intelligence Centro de Investigación y de Estudios Avanzados del IPN — CinvestavBachelor in Computer Science
Computational Science Benemérita Universidad Autónoma de Puebla — BUAP1T31Y80LUUQV kkg359aynw — — 90029376