Francisco José Diego Acosta

Engineered machine learning models utilizing Bayesian Learning techniques to predict short-term electricity demand across regional grids, increasing grid stability by 15%. Pioneered a scalable MLOps platform leveraging MLFlow and GCP Vertex AI, which automated machine learning model deployment and reduced model deployment time by 60%. Engineered a centralized Data Vault architecture on Snowflake,

Authored all learning modules, exercises, and assessments for the Machine Learning course, enabling 80% of graduates to secure data science roles within six months of program completion. Orchestrated a collaborative project with executives from two Fortune 500 firms, synthesizing academic insights into real-world business strategies. Engaged a class of over 30 international students in a pleasant,

Pioneered a multi-language LLM on Busuu using TensorFlow and NLP techniques, pinpointing critical grammatical errors. This initiative raised user engagement scores by 15% through enhanced feedback mechanisms.

Deployed a suite of four machine learning algorithms, leveraging ScikitLearn and AWS, to refine energy consumption predictions. This effort reduced prediction errors by 22% and improved the accuracy of energy-saving tips.

Spearheaded the creation of machine learning models to automate risk assessment for business insurance applications. This led to a 30% reduction in processing time and improved customer satisfaction by 15%.

Orchestrated the creation of a decentralized data architecture, rewarding over 5,000 users with tokens for generating data from mobile devices, boosting data ingestion from 500GB to 1TB weekly. Deployed a scalable fraud detection infrastructure using GCP services, processing over 1 million daily transactions and identifying over 5,000 potentially fraudulent activities with a precision rate exceedi

Architected advanced analytical models for a decentralized data ecosystem across e-commerce, textiles, media, and telecommunications, delivering models to two cross-functional teams with perfect testing scores. Constructed low-latency data pipelines utilizing Kafka and Flink, ingesting over 500,000 events per minute, which decreased fraud detection response times for the e-commerce team.

Pursued a Master of Science in Mathematical Engineering, focusing on advanced mathematical concepts and their applications. Developed strong analytical and problem-solving skills.

Completed a Master of Science in Quantitative Economics, gaining expertise in economic analysis and quantitative methods. Applied statistical and econometric techniques to economic data.

Obtained a Bachelor of Science in Mathematics, building a strong foundation in mathematical principles and problem-solving. Engaged in rigorous coursework covering various branches of mathematics.

Earned a Bachelor of Arts in Economics, studying economic theories, principles, and their real-world applications. Analyzed market trends and economic policies.