About Tyton
Tyton Ecological Intelligence builds AI-first platforms for ecological monitoring at landscape scale.
Our products, TytonAI and TytonIQ, combine machine learning, high-resolution aerial imagery (drone, aircraft, satellite), and cloud analytics to automate vegetation classification, rehabilitation monitoring, weed mapping, and biodiversity assessment across thousands of hectares.
Tyton is creating a globally scalable, enterprise platform used by ecology consultancies and environmental teams to replace manual workflows with repeatable, ML-driven processes. TytonAI is already deployed on major mining operations in Western Australia, and we’re now productising this capability for global use.
We’re a founder-led, fast-growing company focused on building world-leading AI systems that operate reliably at scale. Our engineers work on massive imagery datasets, deep learning pipelines, and real-world ML performance problems rarely found in typical SaaS environments.
The Role
We’re looking for a Senior Data Scientist to help evolve TytonAI, our large-scale computer vision platform for ecological classification.
This is a hands-on senior role combining R&D and delivery. You’ll own core ML systems end-to-end: experimenting with new ideas, turning successful models into production pipelines, and continuously improving performance, reliability, and scale.
You’ll work on deep learning for high-resolution imagery, geospatial pipelines, and cloud workflows supporting enterprise customers globally. Because the platform turns imagery into decision-grade ecological figures, the role uniquely sits at the intersection of four disciplines at once; machine learning (classical ML and deep-learning computer vision on imagery), geospatial / remote-sensing engineering, applied statistics & econometrics, and ML infrastructure. This role suits someone who enjoys shipping real ML products - not just building prototypes.
Key Responsibilities
- Own and extend TytonAI’s PyTorch-based ML codebase (production + R&D) ensuring stability, maintainability, and performance
- Design and run experiments focused on customer impact
- Design, train, and deploy deep learning models for large-scale remote sensing and geospatial classification
- Improve TytonAI’s core computer vision capabilities (segmentation, classification, fine-tuning)
- Translate research ideas into production systems
- Build and optimise pipelines for raster/vector spatial data and large-area processing
- Apply survey / spatial sampling techniques to extract representative, well-stratified tiles/sites for training and validation sets.
- Design evaluation strategies suited to imbalanced and rare classes (rare species, lifeforms), so that reported figures are genuinely trustworthy.
- Identify performance bottlenecks and improve inference speed, memory usage, and cloud efficiency
- Contribute to platform architecture and technical direction as the product scales
- Build tooling for training data, evaluation, and batch inference
- Explore emerging ML techniques, including foundation models and LLMs where relevant
- Work within cloud environments (primarily Google Cloud) and Linux environments
- Maintain documentation and engineering standards
- Mentor junior engineers
Essential Skills and Experience
- 8+ years professional experience across data science, applied statistics, machine learning and deep learning, including a proven track record shipping production ML systems
- Strong PyTorch and Python engineering skills
- Deep-learning computer vision (PyTorch) applied to high-resolution imagery ; training and fine-tuning models for semantic segmentation and pixel/object classification on satellite, aerial or drone data (i.e. modern DL-based CV, not classical-CV-only, tabular- or NLP-only ML).
- Geospatial & remote-sensing engineering: hands-on raster/vector processing, coordinate reference systems, tiling, mosaicking, spatial joins/overlays and polygon geometry — using QGIS and PostGIS.
- Strong classical machine learning alongside deep learning, feature engineering and models such as tree ensembles / gradient boosting, where they outperform or complement neural nets.
- Formal grounding in statistics, probability and econometrics, survey / spatial sampling, uncertainty quantification, controlling for confounders (effects "all else equal"), and evaluation / metric design for imbalanced or rare classes.
- End-to-end MLOps: containerised training and inference (Docker / Kubernetes), CI/CD for ML, experiment tracking, model monitoring and reproducibility.
- Experience maintaining or extending existing ML codebases
- Proven track record shipping production ML systems
- Solid understanding of model training, evaluation, deployment, and optimisation
- Linux + cloud experience (GCP preferred)
- Strong problem-solving ability with real-world data
- Demonstrated experience leading or managing a technical team (ML / statistics / data) while remaining hands-on, setting direction, mentoring, and owning delivery.
- Geospatial / remote sensing experience
- Experience working with large imagery datasets (satellite, drone, aerial)
- GIS and related tool exposure (e.g. QGIS / ArcGIS)
Note that applicants must be based in Perth
What We Offer
- A senior technical role with meaningful ownership of core ML systems
- Fast innovation cycle from idea → experiment → production
- Opportunity to build a globally scalable environmental AI platform
- Competitive salary + flexible hybrid work
- Small, high-calibre team with real product influence
- Perth-based office with a practical, focused, low-ego culture
Pay: $130,000.00 – $155,000.00 per year
Work Location: In person
