Deep Learning Engineer

Deep Learning Engineer About Nanook

Founded in 2008, Nanook is an energy‑focused investment firm built by experienced energy professionals. Since launching our first fund in 2014, we have applied deep fundamental modelling of energy supply and demand to some of the most complex energy markets in the world.

Today, Nanook is a specialist team solving highly complex, real‑world problems in energy trading, building cutting‑edge quantitative and technical solutions where rigorous analysis and disciplined thinking are critical to success.

We are looking for a Deep Learning Engineer to help design, development, and deployment of advanced deep learning models  will tackle some of the most complex problems in global energy markets, applying cutting‑edge neural network and deep learning techniques across diverse and challenging data sets.

The ideal candidate combines deep technical expertise in neural networks and deep learning with practical experience delivering models end‑to‑end, from research to production, in a high‑performance environment. Your work will directly impact trading decisions, turning sophisticated models into actionable insights.

• Advanced degree in a quantitative field.
• 4+ years of hands‑on experience building and shipping deep learning models in production, including ownership of end‑to‑end pipelines.
• Strong proficiency with Python and modern DL frameworks (preferably PyTorch; familiarity with JAX or Tensor Flow a plus).
• Solid understanding of statistics and ML fundamentals.
• Experience of supervised learning and regression problems.
• Proficiency with GPUs and efficient training/inference for out‑of‑memory datasets.
• Strong data engineering fluency: experience working with large, messy datasets; familiarity with distributed compute.
• Ability to communicate clearly with both technical and commercial stakeholders.
• Ownership mindset with a track record of delivering measurable impact under time constraints.
• Ability to recognise when simpler approaches are more appropriate.

• Experience producing probabilistic and distributional forecasts.
• Experience with any of the following:
  • Attention mechanisms and transformer based architectures.
  • Physics informed neural networks.
  • Bayesian neural networks.
  • Spatiotemporal and/or graph‑based modelling.
  • Generative models.
  • Reinforcement learning.
  • Time series forecasting.
• Familiarity with reinforcement learning or generative models would be valuable, although not the focus of the role.
• Experience working with weather data, satellite imagery or spatial data.
• Domain experience in energy, commodities, or macro forecasting.

• Competitive salary with bonus opportunities
• Life insurance of 4x salary
• Healthcare
• 8% Pension contribution
• Gym Membership
• Cycle to work scheme
• Employee assistance program
• Work socials & trips
• Amazing team & working environment