Manuel Enrique Luna Alegria Data Scientist

D

Data Scientist

Jan 2026 - Current (10 months)

Build Stay Time estimation models in Databricks/PySpark, iterating from a linear regression baseline
(V0) toward improvements using mixed-effects models, segmentation strategies, and alternative regression approaches; evaluate performance with MAE, RMSE, wMAPE, Bias, and Accuracy.
∠ Apply non-parametric statistical testing (Wilcoxon signed-rank, rank-biserial correlation) on large
paired-observation datasets (∼88K records) to validate model improvements with statistical rigor.
∠ Manage code repositories on Azure DevOps with a branch-per-environment strategy (dev/qa/prd) and
environment-specific YAML configs; coordinate sprint deliverables, data dependencies with the Data
Office, and stakeholder communications.

C

Data & BI Engineer

Centro Internacional de Agricultura Tropical -CIAT

Mar 2024 - Jan 2026 (1 year 11 months)

Managed end-to-end data pipelines (Microsoft Fabric/Power BI, Python) and maintained database
connections (SQL/Snowflake) to deliver reliable, interoperable data products.
∠ Built and operated web-scraping collectors for Africa-focused programs; curated and indexed content
to ground LLMs; developed lightweight Streamlit frontends; maintained prompt libraries and evaluation checks.
∠ Improved data governance (lineage, definitions, access control) across digital products for Alliance
and CGIAR initiatives; established standards and documentation.

Fraud Prevention Lead

Rappi

Feb 2023 - Mar 2024 (1 year 2 months)

Led the end-to-end fraud-risk modeling strategy across 9 LATAM countries, designing and deploying
supervised ML models (logistic regression, decision trees, Random Forest, XGBoost) for transaction,
account-takeover, and promo-abuse detection.
Engineered behavioral, device, velocity, and network-based features in Python; built reusable feature
pipelines and ensured reproducibility across training and scoring environments.
Operated SQL/Snowflake pipelines for model training, batch scoring, and back-testing; orchestrated
retraining cadences and champion–challenger experiments to mitigate concept drift.
Monitored model health and fraud KPIs (AUC, KS, PSI, precision–recall, false-positive rate, paymentapproval

T

Data Quality Analyst

TQ Confiable - Tecnoquímicas

May 2022 - Feb 2023 (10 months)

Built and maintained Power BI dashboards and KPI frameworks (yield, defects, stability, throughput,
plan adherence), standardizing data models and definitions across sites for consolidated executive
reporting.
Implemented Statistical Process Control (SPC) and capability analysis (control charts, process sigma,
Cp/Cpk), enabling early deviation detection and structured root-cause analysis.
Partnered with the Technical Quality Control Manager to deliver analytics and reporting across multiple countries and the company’s four production units, aligning quality, operations, and management on shared metrics.
Developed and deployed analytics dashboards to interpret information and visualize key metrics and
trend

Professional in statistics

Clínica de oftalmología de Cali

Aug 2021 - May 2022 (10 months)

- Carry out the analysis of the Clinic's information for decision-making by the management, developing various weekly and monthly reports for the board of directors, as well as providing support in the areas of Surgery, Outpatient Services, and Human Resources in data management for the generation of information that enables informed decisions to be made.
- All of this involves implementing statistical techniques to create KPIs for eacharea in Power BI for Clinic Management.

Teacher Assitant

Pontificia Universidad Javeriana

Feb 2021 - Jul 2021 (6 months)

Support in the teaching of Applied Statistics, Statistics for decision making, Statistics I, Statistics II, and related subjects in university programs such as Business Administration, Engineering, Psychology, Biology, and others. The main focus was on teaching statistical software like R, R-Studio, Minitab,SPSS, and Excel functions

Statistics Intern

Grupo Manuelita

Aug 2020 - Feb 2021 (7 months)

- Analysis of experimental designs in the area of Agricultural Technology.
- Construction of a Development Index for the variety from Sugar Cane.
- Training for area personnel in data analysis in experiments.
- Realization ofestimated forecasts in Sucrose.
- Characterization of productivity according to variety of sugarcane.
- Performance of multivariate analysis (ACP-ACM) to observe the behaviorof each variety of sugarcane by type of soil and variety of environmental conditions, characterizing the best yield of sucrose for each type of variation.

Teacher Assitant

Pontificia Universidad Javeriana

Feb 2020 - May 2020 (4 months)

Support in the teaching of Applied Statistics, Statistics for decision making,Statistics I, Statistics II, and related subjects in university programs such asBusiness Administration, Engineering, Psychology, Biology, and others. Themain focus was on teaching statistical software like R, R-Studio, Minitab,SPSS, and Excel functions

Research Assistant

Universidad del Valle (CO)

Mar 2018 - Jun 2018 (4 months)

As a statistics student, support the development of a measurement instrument that allows identifying the perception of students with disabilities at Universidad del Valle.

Research Assistant

Universidad del Valle (CO)

Feb 2018 - Dec 2019 (1 year 11 months)

Manuel Enrique Luna Alegria

About

Detail

Timeline

Résumé

Manuel Enrique Luna Alegria

About

Detail

Behavioral traits