Time Series Researcher

Applied Computing is a company focused on building a physics-informed foundation model for energy operations. The Time Series Researcher will design, validate, and deploy foundational time-series models, ensuring their application in real-world scenarios and contributing to the company's mission of sustainable energy operations.

Responsibilities:

• Design core time-series architectures supporting:
• Forecasting
• Classification / anomaly detection
• Optimisation & control-adjacent tasks
• Explore and select appropriate objectives examples:
• Probabilistic losses
• Generative formulations
• Reinforcement-learning-inspired objectives where appropriate
• Develop hybrid approaches that blend:
• Classical statistical models
• Deep learning architectures
• Physics-based constraints
• Integrate domain physics into learning systems, including:
• Conservation laws
• Process constraints
• Differential-equation-based priors
• Improve generalisation, interpretability, and extrapolation beyond training regimes
• Ensure models respect physical feasibility in production settings
• Design uncertainty-aware models (Bayesian, ensemble, hybrid)
• Quantify confidence under:
• Sensor drift and failure
• Regime change
• Sparse or delayed ground truth
• Ensure outputs are usable by operations and engineering teams, not just statistically elegant
• Containerise and deploy models using Docker on AWS / Azure (EKS, ECS, SageMaker)
• Build or integrate CI/CD pipelines for:
• Training
• Evaluation
• Rollout and rollback
• Automated retraining triggers
• Define rigorous back-testing and evaluation protocols
• Build automated benchmarking pipelines across datasets, regimes, and failure modes
• Compare against classical baselines and modern deep-learning approaches
• Ensure claims are defensible to customers, partners, and internal stakeholders
• Deep understanding of Orbital’s data, domains, and production constraints
• Contribution to at least one core time-series model or experimental track
• Clear ownership of a modelling problem with defined success metrics
• One or more foundational models running reliably in production
• Demonstrable improvements in:
• Forecast accuracy
• Robustness under faults
• Uncertainty calibration
• Models actively used by downstream agents and optimisation layers
• Benchmarking standards adopted across the research team