This role is with one of Dex’s trusted partner companies. We work closely with their teams to truly understand their culture, goals, and what they’re looking for, so we can match you with the right opportunity and give you context about the role before you commit to a process.
The role
Most AI companies chase the same shiny problems. This one doesn’t. They’re building an AI operating system for industries modern software has ignored for decades, where data is fragmented, incomplete, and complex. Think real‑world impact, not just another consumer app. They’re backed by leading VCs, and their founding team are seasoned AI researchers.
You’ll join a small, high‑calibre team, working directly with the founders. This isn’t a pure research role; it’s applied science. You’ll take messy, real‑world data from legacy systems and transform it into AI‑driven products, from forecasting to optimisation. Expect deep data exploration, rigorous experimentation with LLMs, and models that actually ship. This isn’t about academic papers; it’s about building production‑ready AI that delivers tangible business value.
The work
- Wrangle large, fragmented datasets from legacy systems, transforming raw data into usable features for AI models.
- Design and execute rigorous experiments, establishing proper baselines and statistical evaluation for new models and features.
- Build, deploy, and iterate on applied ML models for forecasting, optimisation, and recommendation across core business functions like sales and inventory.
- Explore and apply cutting‑edge LLM techniques, from fine‑tuning to prompt engineering, to unlock new product capabilities.
- Collaborate directly with engineering to productionise research, ensuring models are robust, scalable, and trustworthy.
What You Bring
- A Master’s or PhD in a quantitative science (e.g., ML, physics, neuroscience, mathematics). A strong scientific background with self‑taught ML is welcome.
- 4+ years of hands‑on experience in data science, ML engineering, or applied research (or 2+ years post‑PhD).
- Proven ability to design and execute statistically rigorous experiments, including proper baselines and controls.
- Expertise in Python and core DS/ML libraries (e.g., pandas, scikit‑learn, PyTorch, Hugging Face), with a deep understanding of evaluation metrics.
- A genuine drive to apply and experiment with LLMs in practical, business‑focused product contexts.
#J-18808-Ljbffr…