Data Scientist - Time Series Forecasting
Confirmed on the employer's own hiring board on Oct 27, 2026. First seen by Alion on Oct 26, 2026.
Description
Fetcherr builds responsible AI that turns market complexity into measurable profit growth. At its core is the Market Model, a proprietary AI model that delivers granular demand predictions and real-time decision intelligence for commercial teams. Built on a glass-box architecture, it uses market data, not personal data, with full transparency into logic and outcomes. First deployed in global aviation, the technology is industry-agnostic, and Fetcherr delivers a consistent average profit uplift of 7%.
Fetcherr Labs is a new unit that takes this engine to mid-size companies, run by agents and delivered inside the tools customers already use, such as Excel and Copilot. Labs is small, flat and senior, and builds with AI agents from day one. You will work directly with the Market Model Lead and report to the Chief of Labs.
The role
We are looking for a Demand Modeling Scientist who owns demand models end to end: from a new customer's raw history to a model running in production, measured against real outcomes. You will teach the engine how demand behaves in new industries, starting with aftermarket parts, and prove the uplift through live A/B tests with our design partners.
What you'll do
- Develop and implement state-of-the-art econometric and machine learning models for demand forecasting.
- Own the full machine learning lifecycle: data exploration, feature engineering, training, validation, deployment, monitoring, retraining and retirement.
- Conduct research and experimentation to improve accuracy, robustness and scalability, and adapt models to new industries and customer data.
- Design how models are measured in production, including A/B tests that show the uplift for each customer.
- Build with AI coding agents as your everyday way of working: agents write and test much of the code, and you design, review and decide.
- Work with data and platform engineers to put models into production on Azure, and keep them explainable and reliable.
- Explain complex findings clearly to non-technical stakeholders, including customers and executives.
Requirements
What you bring
- 5+ years of hands-on experience in data science and machine learning, with a proven record of turning models into business outcomes.
- Proven experience taking models through the full lifecycle into production: versioning, testing, reproducibility, deployment, monitoring and retraining.
- Expertise in time-series forecasting, ideally deep-learning based, preferably in demand prediction or related areas.
- Feature engineering, feature importance testing and explainability experience.
- Proficiency in Python and its ML and data stack (PyTorch or TensorFlow, Pandas or Polars, NumPy, Scikit-learn, SQL).
- Hands-on use of AI coding agents (such as Claude Code or GitHub Copilot) in your daily development, and a strong view on where they help and where they don't.
- Master's or PhD in Computer Science, Machine Learning, Statistics, Engineering or a related field.
- Excellent communication and collaboration skills.
Nice to have
- Experience with pricing, revenue management, retail or B2B distribution.
- Experience with experimentation and causal measurement (A/B testing, uplift).
- Experience with cloud ML platforms (Azure ML or Microsoft Fabric preferred; GCP or AWS also relevant).
- Familiarity with workflow orchestration in general (any modern orchestrator).
- Publications in top-tier peer-reviewed ML venues (ICLR, ICML, NeurIPS) or strong results in data science competitions.
If you want to own demand models end to end, build with agents, and bring a proven AI engine to a whole new market, we'd love to hear from you.