Daniel Tobon Collazos
About me
I am a Platform Engineer focused on cloud infrastructure, automation, and developer platforms, with a strong background in ML Platform and MLOps. I have experience designing and building reliable cloud-based systems using Kubernetes, Terraform, Helm, Docker, CI/CD, and cloud-native services across Google Cloud and Azure.
Most recently, I worked as an MLOps Engineer at SmartJob, collaborating with Falabella Tech on production ML platform capabilities. My work included Kubernetes-based services, deployment automation, workflow orchestration, observability, and cloud-native infrastructure.
At AB InBev, I contributed to the modernization of a centralized ML platform across Azure and Databricks, working on infrastructure automation, Terraform, cloud networking, Kubernetes, Helm, CI/CD, reusable engineering components, and standardized development and deployment workflows.
Previously, I worked as a Machine Learning Engineer at Xmartlabs, collaborating with Levi Strauss & Co. on production ML systems and cloud-native infrastructure. I worked with Kubernetes, Vertex AI, Docker, GitHub Actions, Workload Identity Federation, artifact management, secrets management, and deployment automation.
I obtained my BSc in Mechatronics Engineering from Universidad Autónoma de Occidente, where I developed a robotics perception system to estimate geometric features in trees as my degree project.
I grew up in Santiago de Cali, Colombia. I am passionate about building reliable systems, automating engineering workflows, and continuously learning technologies that improve how teams build, deploy, and operate software.
Core Technologies
Platform & Infrastructure
Kubernetes
Terraform
Helm
Docker
Linux
Cloud
Google Cloud
Azure
Cloud Networking
IAM / RBAC
Secrets Management
Automation & Delivery
GitHub Actions
CI/CD
Infrastructure as Code
Python
Bash
ML Platform
Vertex AI
Databricks
Airflow
MLflow
ML Pipelines
Publications & Research
A photogrammetric system for dendrometric feature estimation of individual trees
Daniel Tobon Collazos
IEEE Colombian Conference on Robotics and Automation (CCRA)
.
Work Experience
MLOps Engineer – SmartJob
Collaborated with
Falabella Tech
on cloud-based platform capabilities for production machine learning workloads.
Focused on Kubernetes, deployment automation, workflow orchestration,
observability, and reliable production systems.
* Built Kubernetes-based services using Helm, FastAPI, Ingress, and CI/CD
* Designed Airflow workflows for automated data processing and ML pipelines
* Integrated Vertex AI, BigQuery, PostgreSQL, and cloud-native services
* Implemented monitoring and alerting with Grafana, Loki, and Microsoft Teams
* Built local development environments using Docker Compose
Machine Learning Engineer – Xmartlabs
Collaborated with
Levi Strauss & Co.
on production ML systems with a strong focus on cloud infrastructure,
containerization, deployment automation, and secure CI/CD workflows.
* Deployed containerized workloads on Kubernetes and Vertex AI
* Reduced Docker image size from 16 GB to 8 GB
* Reduced Vertex AI pipeline execution time from 16 min to 5 min
* Migrated CI/CD from Jenkins to GitHub Actions, reducing deployment time from 2h to 5 min
* Implemented Workload Identity Federation for secure cloud authentication
* Implemented artifact management, secrets management, CodeQL checks, and alerting
Senior Machine Learning Engineer – AB InBev
Contributed to the modernization of a centralized ML platform on Azure and Databricks,
focusing on infrastructure automation, platform standardization,
reusable engineering components, and developer enablement.
* Automated Azure and Databricks infrastructure using Terraform
* Designed Development, Staging, and Production workspace environments
* Built reusable Terraform modules and YAML-driven Databricks Job automation
* Developed reusable Kubernetes deployment tooling using Helm and GitHub Actions
* Implemented integration testing, monitoring, alerting, and standardized secrets management
* Received internal recognition for reusable DevOps and MLOps components
Machine Learning Scientist – AI Turing
Worked on computer vision and deep learning workloads while gaining hands-on experience
with cloud-based ML infrastructure and production-oriented ML workflows.
* Modernized deep learning pipelines from TensorFlow 1 to TensorFlow 2
* Developed object detection and classification workloads
* Implemented data augmentation and custom evaluation metrics
* Used Vertex AI, Cloud Functions, and Cloud Storage for ML workflows
* Containerized training and inference workloads across AWS and Google Cloud
* Researched MLOps approaches for production ML systems
Research Engineer – IMAR
Worked on R&D projects involving software development, embedded systems,
computer vision, IoT, and sensor-based applications.
* Developed computer vision and industrial IoT prototypes
* Built interfaces for Intel RealSense cameras using PCL and ROS
* Developed embedded applications using ESP32 and ESP-IDF
* Developed C/C++ software and CMake-based projects
* Integrated hardware and software components for sensor and RFID systems
Selected Projects
Density-based spatial clustering of applications with noise (DBSCAN) is a data clustering algorithm
This project is a photogrammetric system for dendrometric feature estimation of individual trees. The purpose of this project is to do a 3D reconstruction of an individual tree using Open Multiple View Geometry (openMVG) and get dendrometry estimation (diameter at breast height (DBH), tree crown height, total tree height, crown volume, morphic factor and percentage canopy missing) of a stem tree
C++ application to convert pcd file, ply file, txt file or xyz point cloud to MESH representation (Gp3).