Rock Robotic
Director of Technology
Dec 2020 - July 2024
Location: Remote
Management
- Managed AWS infrastructure for processing
- Led a team of worldwide developers in the creation of new products and support of existing products through Agile practices and team rituals
- Support customer success with triaging and resolving customer tickets
- Managed large S3 Instance Managed 415.1 TB Total storage over 15.1 M Object count with a 28.9 MB Average object size
- ~1200 tickets Managed and created with a 90% completion rate across a 5 person dev team
- Work cross departmental to understand business requirements and deliver effective solutions
Backend
- Developed APIs for cloud-based LiDAR data processing and created a comprehensive platform for managing, processing, and visualizing LiDAR data
- Implemented licensing model that enabled in-house hardware processing and catalyzed the launch of new product lines
- Architected and engineered a message queue architecture for asynchronous LiDAR data processing and co-maintained a Python package for specialized data processing
- Developed CI/CD to build, distribute and update Electron app
- Utilized AWS Cloudfront to serve over 3000GB/month with 12 million requests
Frontend
- Developed Electron based desktop application integrating React and NodeJS (ROCK Desktop); in-housing licensing and revenue from products.
- Designed and implemented custom UI/UX, based on MUI, for processing, managing and sharing data
- Implemented bespoke, highly performant rendering of large datasets using WebGL and ThreeJS
- Created uploader for large files utilizing AWS S3 Multipart handling. Currently ~15M have been uploaded, with sizes ranging from 100 KB -> 100GB.
- Created customer dashboards to manage projects and folders
- Designed near real time messaging system to allow for customer feedback during long processes; reducing customer support tickets by 65%
- Integrated APIs with cloud services and desktop applications including: CRUD operations, authentication, uploading large data, licensing, purchasing, and hardware configuration management
- Utilized custom WebGL shaders for real-time viewing of point cloud classification and edits
- Opened new lines of business through integrating OpenLayers map, leading to increase in revenue
- Reduced deliverable turn around time by 75% through automated ML classification workflows, in-cloud editing, and enhanced data visualization for easier QA/QC.
- Managed AWS EC2 and S3 instances to support large-scale secure data operations
- Contributed to the development of the COPC file format, enhancing data handling efficiency
- Used Web Workers and ThreeJS to calculate and visualize volume from millions of points
- Enhanced data visualization capabilities by developing a map display integrating contours, trajectory, photos, and more, significantly improving product functionality