Denzel Wamburu Email & Phone Number

- Designed and optimised high-performance data pipelines to process satellite data for climate monitoring and carbon footprint modelling, significantly reducing overall processing time.- Partnered with data scientists to embed machine learning models for extracting climate indicators from multispectral and hyperspectral imagery, leveraging computer vision and sensor fusion techniques.- Led the creation of a centralised data repository, converting unstructured satellite data into a normalised, queryable format; implemented advanced indexing strategies to improve retrieval times across API services.- Developed a specialised toolkit for feature extraction from satellite imagery, enhancing model interpretability and supporting more accurate climate predictions.- Built interactive, user-friendly interfaces to display and analyse spatial data, prioritising accessibility and functionality for internal ML models. Utilised WebGL, Three.js, Mapbox GL, and Leaflet for rendering complex 3D geospatial visualisations directly in the browser, enabling users to explore satellite data interactively.

Nairobi, Kenya

- Designed and implemented data pipelines for collecting, transforming data for connected farm management system from various sources including external & internal APIs, hardware sensors and data lakes. - Collaborated with domain experts, agronomists, and data scientists to design models powering recommendation systems to address yield prediction, crop disease detection, and precision agriculture.- Integrated ML models into production systems and pipelines over AWS.- Developed and maintained scripts for data preprocessing, model training, and evaluation, optimizing the end-to-end machine learning workflow. Increased model accuracy on the edge by 40% by utilising swarm intelligence.- Designed and Implemented unit and end to end testing for backend, data pipelines and frontend.- Technologies used: Spark, FastApi, Clickhouse, Kafka, PostgreSQL, Metaflow, PyTorch, Tensorflow.

Kenya

- Led the migration of on-premises legacy systems to Azure, resulting in a 40% reduction in infrastructure costs and improved system performance.- Designed and implemented highly available, scalable and automated infrastructure on Microsoft Azure for multiple projects, leveraging services such as Azure Virtual Machines, Azure Kubernetes Service (AKS), Azure Functions, and Azure App Service.- Implemented and managed Azure DevTest Labs to provide development and testing environments on-demand, reducing infrastructure provisioning time by 30%.- Implemented automated scaling policies for Azure AKS, dynamically adjusting resources based on demand, resulting in improved application responsiveness and cost optimization.- Technologies used: ELK Stack, Kubernetes, Terraform, Selenium, Jenkins, Ansible, RobotFramework.

Nairobi County, Kenya

- Designed, developed, and maintained high-performance, scalable embedded software applications for real-time advertising quotas trading platform and internet usage based billing based on tokens & subscriptions cycles. - Collaborated with hardware and software teams to define sensor requirements and specifications for new systems.- Conducted performance testing and validation of sensor fusion algorithms through simulations and real-world testing.- Actively participated in agile development methodologies, including daily stand-ups, sprint planning, and retrospective meetings, to enhance team collaboration, productivity, and delivery timelines.- Technologies used: QT, ROS, MQTT, LVGL, TinyUSB, RTOS, Embedded Linux, OpenAPI.

London, United Kingdom

Developed Computer Vision model to aide in dermatology diagnosis and recommendations. Integrated models with mobile app and data pipelines.

Remote

Contributed to open source development of command line tools for continuous integration, unit tests, cloud deployments & over the air updates to on/off-field hardware devices.