Résumé

Profile

Industrial Engineer (Javeriana) with ten years of experience —from process engineering to data and AI— turning operational data into business decisions, across banking, transport, logistics and healthcare. Today I lead AI adoption and governance: the strategy and the ISO/IEC 42001 implementation. Before that I designed from scratch the data strategy and platform of an AI-agent startup —Microsoft Fabric and a replicable process to design and monitor agents— and built BI and predictive analytics in banking. Microsoft-certified as a Fabric Analytics Engineer (DP-600), on the way to AI-103 and AI-300. Want to know anything else? Ask this résumé's chat.

Career

1. Analytics Professional — Fundación CTIC

   2025 — today
   • I lead the institutional AI strategy under UNE-ISO/IEC 42001:2025: 12 opportunities identified, 7 cases evaluated and 3 prioritized.
   • 42 analytics products in Power BI for administrative and clinical leaders: 20 leaders of 15 processes and about 75 users.
   • Institutional data governance and quality, under Colombia's habeas data regime and Law 1581.
   • Process-based control dashboards and improvement plans: close to 60% less effort preparing information.
   • 23 instruments of the AI management system, 8 finished, and an ISO 42001 expert agent to review them.

2. Data Strategy Lead — Vesting — automation-agents startup

   2023 — 2025
   • Designed and built from scratch the data ecosystem for AI-agent analytics on Microsoft Fabric (big data + data warehouse + distributed processing): 12 clients, 120 tables and 1,000 events per day.
   • Defined the data governance to integrate client and system data, with each client isolated in its own workspace.
   • Built real-time monitoring for the agents in production: up to 23 monitored at once.
   • Documented the replicable 11-stage core process used to build the 27 agents in the inventory.

3. Senior Analytics and Reporting Analyst — Banco Pichincha

   2023
   • Led a 5-person BI team: dashboards adopted by 50+ users, driving +25% in decision-making.
   • ETL process optimization in Power Query: −35% in analysis times.
   • Churn, delinquency and risk models with scikit-learn: >90% accuracy and +35% in predictions.
   • Semantic model optimization with DAX Studio and Tabular Editor.
   • Co-led the area's data governance.
   • Analytics training program for 12 professionals: +20% productivity in preparing and using information.

4. Post-Operational Analysis — C&M Consultores (TransMilenio)

   2021 — 2022
   • Drove BI adoption across operations: +35% efficiency with 25+ key users.
   • ETL that unified five heterogeneous sources (fare collection, fleet and GPS, scheduling, incidents and user complaints): +70% in analysis precision and speed.
   • Automation scripts with validations and visible exceptions: −40% in repetitive tasks.
   • Strategy sessions with SITP leadership: +25% in indicators.
   • Demand forecasting with scikit-learn by route and time slot, ten months in use: +20% system performance.

5. Information Systems Analyst — Cafam

   2020 — 2021
   • Led a 20-person team —14 from Cafam and 6 from Oracle— through six months of WMS implementation testing: −25% errors and +15% operational efficiency.
   • Modeled the process in BPMN with Bizagi and simulated dispatch in FlexSim before testing.
   • Operational-control BI: +50% precision for 15+ users.
   • VBA integrations with the WMS: +15% automation and −50% data errors.
   • SQL data quality with an 80-20 reconciliation by medicine: +20%.

6. Operations Analyst — C&M Consorcio (TransMilenio)

   2018 — 2020
   • Data-driven supervision of about 150 routes from 10 concessionaires, cross-referencing five sources: fare collection, fleet and GPS, scheduling, incidents and user complaints.
   • Performance dashboards and reports —two weekly and one monthly— traceable down to the operational event.
   • Automation with Excel, VBA and SQL: at least −40% in processing time.
   • The system's statistical memory in SQLite: the operation's history, week by week.
   • Audit of compliance with operating protocols, with indicators carrying contractual consequences.

7. Operations Coordinator — Ceinfes

   2017 — 2018
   • KPIs for scheduling, logistics and digitalization of mock exams for more than 100 schools a year; weekly reports to the board.
   • Coordination of about 40 people across three fronts and some 50 exam proctors.
   • Balancing the digitalization line: about 250 sheets per shift across two scanning stations.
   • A VBA macro that optimized teacher scheduling with explicit constraints.
   • Transition to process-based management with BPMN, Kanban and Scrum.

8. Process Engineer — Inglopres

   2016 — 2017
   • Led the implementation of the Odoo ERP to integrate the processes of a heavy-machinery rental and sales company, with a fleet of about 120 units.
   • Designed in SQLite and SQL the analysis databases the operation did not have.
   • Work study with fatigue allowances for sustainable time standards.
   • Supply chain and quality assurance under ISO 9001:2015.
   • A team of 12 people with 95% client satisfaction.

Projects

• A data platform for AI agents — Vesting (2023–2025)

  An automation-agents startup with no data infrastructure: I designed and built from scratch its Microsoft Fabric ecosystem (big data + data warehouse + distributed processing), the governance to integrate client and system data, real-time agent monitoring, and the replicable core process to design and ship AI agents. The exact bridge between analytics engineering and AI engineering.

  View case study→

• BI the business actually uses — Banco Pichincha (2023)

  Dashboards adopted by 50+ users (+25% in decision-making), ETL times cut by 35%, predictive models above 90% accuracy (scikit-learn), co-led governance and a training program that lifted productivity by 20%.

  View case study→

• Data that moves a city — TransMilenio / C&M (2021–2022)

  BI adoption (+35% efficiency, 25+ key users), an ETL that unified heterogeneous sources (+70% in precision and speed), demand forecasting with scikit-learn (+20% system performance) and strategy sessions with SITP leadership (+25% in indicators).

  View case study→

• Leading the quality of a WMS — Cafam (2020–2021)

  A team of 20 through implementation testing: −25% errors and +15% efficiency; control BI (+50% precision, 15+ users) and VBA integrations (−50% data errors).

  View case study→

• Analytics in healthcare — Fundación CTIC (2025–today)

  Analysis and visualization models for administrative and clinical leaders, institutional data governance and quality, and process-based control dashboards.

  View case study→

• Supervising a city with data — C&M Consorcio / TransMilenio (2018–2020)

  Data-driven supervision of TransMilenio's operation: about 150 routes from 10 concessionaires, five sources cross-referenced (fare collection, fleet and GPS, scheduling, incidents and user complaints), performance dashboards and reports traceable down to the event, automation with Excel, VBA and SQL (at least −40% processing time) and the system's statistical memory.

  View case study→

• Running an operation with indicators — Ceinfes (2017–2018)

  Coordinating three fronts —scheduling, digitalization and logistics, about 40 people— for assessments at more than 100 schools a year: KPIs by area, balancing the digitalization line, a VBA macro for scheduling, weekly reports to the board and the transition to process-based management with Kanban and Scrum.

  View case study→

• Building the data no one had — Inglopres (2016–2017)

  First job as Process Engineer at a heavy-machinery rental and sales company (about 120 units): implementation of the Odoo ERP, analysis databases in SQLite and SQL, work study with fatigue allowances, supply chain under ISO 9001:2015 and a team of 12 people with 95% client satisfaction.

  View case study→

Education

• Industrial Engineering, emphasis in Data Analytics Intelligence — Pontificia Universidad Javeriana · Bogotá

  2009 — 2016

  Process optimization, data analysis and decision support, integrating engineering principles with advanced analytics. Modelling and prediction of industrial phenomena with SQL, Python and Power BI.

• Undergraduate studies in Industrial Design — Pontificia Universidad Javeriana · Bogotá

  2011 — 2016

  Product and systems design with a technological, market-driven approach, focused on environmental sustainability and cultural impact.

• Intensive English course and IELTS certification — Cambridge International College · Melbourne, Australia

  2013 — 2014

  English B2: reading, listening and technical writing; professional discussions and analytical reports.

Certifications

• Microsoft Certified: Fabric Analytics Engineer Associate (DP-600) (2024)
• Microsoft Certified: Azure AI Apps and Agents Developer Associate (AI-103) (In progress)
• Microsoft Certified: Machine Learning Operations Engineer Associate (AI-300) (In progress)
• IBM: Applied Data Science with R (2024)
• IBM: Data Science Professional Certificate (2022)
• IBM: Python for Data Science (2022)
• IBM: SQL for Data Science (2022)

Skills

• AI & ML: Azure AI · RAG & agents — being built in the open (see the showcase) · n8n · PyTorch · TensorFlow · Scikit-Learn · NLP
• Data platform: Microsoft Fabric · Direct Lake · RLS (row-level security) · ETL · Data modeling · SQL · Power Query · DAX · Google Cloud — exploring
• BI & decision-making: Power BI · Data storytelling · KPIs · Data governance & quality · Shiny (R) · Tableau · Looker Studio
• Engineering: Python · Git/GitHub · CI/CD — this site runs on it
• Processes & simulation: BPMN (Bizagi) · Discrete-event simulation (FlexSim) · Time study and line balancing · Kanban and Scrum
• How I work: Leading teams of up to 20 people · Communication with boards and senior leadership · Process-based management and agile methods · User training and adoption · Reproducible documentation and handover · Multidisciplinary teamwork · Professional English (B2)