H
HudHud Maps builds mapping and navigation infrastructure using a machine-learning-first stack—TensorFlow, PyTorch, OpenCV, PostGIS, and PostgreSQL—focused on road detection, object tracking, and semantic segmentation from mobile video. The engineering and data hiring mix (8 of 14 active roles) reflects their core challenge: scaling ML model integration and automating map creation to improve data accuracy and coverage across the region.
Tech Stack 55 technologies
Core StackTensorFlow PyTorch Python C++ scikit-learn ArcGIS Apache Spark PostgreSQL ClickHouse Tableau Looker Power BI Jenkins GitLab CI/CD Java Kotlin Swift Go OpenCV GDAL PostGIS QGIS R GeoPandas Apache Superset JupyterHub SQL Appium XCUITest Espresso+25 more
What HudHud Maps Is Building
◆Challenges
- Improving map data accuracy
- Scaling machine learning models
- Integrating ml models into existing systems
- Scaling our platform
- Performance bottlenecks
- Automating map creation
- Improving coverage and freshness
- Cost efficiency
- Improving reviewer productivity
- Converting pipeline to contracts
▲Active Projects
- Machine learning algorithm development for mapping
- Model training for road detection and semantic segmentation
- Integration of ml models into mapping systems
- Feature extraction from mobile mapping videos
- Object detection and tracking for mobile mapping
- Nlp and llm-based solutions for real-world applications
- Scalable ml pipelines
- Ml model integration into hudhud platforms
- Poi portal & map gen tooling roadmap
- Ai feature prioritization & success criteria
Hiring Activity
15 roles · 8 in 30d
Department
Seniority
Company intelligence
