Dean Shabi

Dean Shabi · AI Founder & Senior Data Scientist

I build AI-powered forecasting, pricing, and automation systems for UK/EU renewable energy markets.

Partnering with energy traders, grid operators, and founders to ship measurable uplifts in accuracy, risk control, and revenue resilience across renewables portfolios.

6+ years
Building AI for renewable energy trading & grid ops
↑57%
Wind forecasting accuracy uplift delivered
£50/MWh
Cost savings unlocked through supply matching

Profile

Data scientist, energy expert, AI engineer, and AI founder.

Physics and electrical engineering roots, followed by 6+ years building production AI systems for UK/EU renewable energy trading, pricing, and grid operations.

I partner directly with founders, traders, and operators to turn ambiguous market problems into measurable forecasting, optimisation, and automation products.

I lean in as a hands-on lead—setting strategy, designing systems, and shipping alongside teams with clear governance and post-launch rituals.

Currently leading founder work on Bloome AI to bring calm, high-signal job search automation to market.

Where I create value

  • Renewable energy forecasting and weather-aware modelling for trading and grid decisions.
  • Risk-based pricing and optimisation engines with clear KPIs and regulatory fit.
  • Operational analytics and automation that teams can maintain after handoff.

Energy market fluency

Ofgem regulationsREGOsBalancing mechanismTrading frameworks

Current Work

Recent products and outcomes.

Spotlighting my new product and recent delivery wins that shaped this year's work.

Founder · Current

Bloome AI

Visit site

AI-driven job matching that delivers five perfect roles every 09:00 Morning Drop.

  • Morning Drop curates investor-grade roles daily, turning job search into a one-minute ritual.
  • Apply Pro unlocks one-click submissions with human-sounding outreach and tailored résumés.
  • Adaptive calibration engine learns from a few approvals; Hands-Free (Beta) applies transparently under supervision.

Mission

Make job searching effortless with human-centred automation that restores calm.

Select delivery

  • tem.energy · RED pricing engine

    Automated portfolio pricing that balances growth with risk controls for UK renewable traders.

  • AmpX · Grid intelligence

    Deployed forecasting platforms and battery degradation models powering real-time decisions.

  • Datamole · Industrial ML

    Delivered predictive maintenance, anomaly detection, and time-series automation across IIoT fleets.

Selected Work

AI-powered projects for energy, trading, and beyond.

Filter by focus area to see the forecasting systems, pricing engines, optimisation models, and tooling shipped with founders and cross-functional teams.

High-Accuracy Forecasting Models
Renewable Energy AI

High-Accuracy Forecasting Models

Led the development of long-term load, generation, and price forecasting models—including PV generation, battery degradation, and price forecasting. These models power core product features, support trading decisions, and reduce balancing costs.

↑57% forecasting accuracy uplift

Energy-TechForecastingML
View project →
Portfolio Pricing Engine
Energy Trading

Portfolio Pricing Engine

Built a risk-based pricing optimisation algorithm that enables energy companies to grow rapidly while balancing risk and competitiveness—a critical edge in volatile energy markets.

↓ portfolio VaR while scaling revenue

Energy-TechOptimizationRisk Analysis
View project →
Exempt Supply Matching Algorithm
Energy Trading

Exempt Supply Matching Algorithm

Created a unique matching algorithm under UK regulations that optimally pairs SME energy consumers with local generators, unlocking ~£50/MWh in cost savings—generating millions in new revenue.

≈£50/MWh cost savings unlocked

AlgorithmEnergyOptimization
View project →
EV Driver Behaviour Simulator
Grid Optimization

EV Driver Behaviour Simulator

Developed a modular, agent-based simulator to model heterogeneous electric vehicle charging behaviours across diverse driver populations, with an interactive Streamlit dashboard for scenario testing.

Validated rollout and tariff decisions

SimulationEnergyPython
View project →
UK's Wind Energy Generation ⚡💨
Renewable Energy AI

UK's Wind Energy Generation ⚡💨

Developed an advanced machine learning model for day-ahead wind power forecasting with remarkable improvements in accuracy, exceeding existing solutions by up to 57%.

↑57% day-ahead accuracy vs incumbent

EnergyForecastingML
View project →

More projects

  • Robot Failure DetectionImplemented anomaly detection algorithms to identify and predict robot failures in automotive manufacturing.
  • Reducing Fruit Waste with MLDeveloped computer vision models to detect fruit ripeness and reduce waste in the food supply chain.
  • AI Agents for Stock Analysis 🤖Created an Equity Analyst Copilot using CrewAI agents framework to revolutionize stock analysis, leveraging specialized AI agents with access to SEC filings, financial documents, and latest stock values.
  • Open Source MLOps Tools ⚙️Comparative analysis of open-source MLOps tools including TensorFlow Serving, MLflow, AWS SageMaker, Seldon Core, and BentoML for efficient deployment and serving of ML models.
  • Streamlit Advanced Guide 📊🤖Exploration of advanced Streamlit features and tips for maximizing its utility, covering topics like multipage apps, logo integration, third-party libraries, data caching, and query parameter usage.

Testimonials

Feedback from teams I've partnered with.

Product leadership

Turns complex roadmaps into clear, actionable plans.

Dean's work has been consistently impressive — his ability to structure complex information at the right level of depth, focus on insights over raw metrics, and adapt quickly to feedback really stood out. His first project was managed with care and clarity, and I always felt well informed throughout. He's shown great initiative by improving his approach between updates and being open to more efficient async workflows. All in all, he's doing brilliantly — I've been consistently impressed with his execution and thoughtfulness.

Imogen

Product Manager, tem.

Data & analytics leadership

Makes data-heavy work understandable.

Dean has a real talent for clarity — the way he presents information makes even complex, data-heavy content easy to follow. It's clear he puts in the effort to make things digestible for people without a statistical background, which is incredibly valuable in cross-functional teams. His communication style is clear, inclusive, and well considered.

Ross

Head of Data, tem.

Executive sponsor

Trusted operator for executive-level problems.

Dean has been a super member of our team, consistently impressing me with every interaction. His smart and thoughtful approach to complex problems set him apart. His intelligence combined with his collaborative spirit makes him an exceptional asset to any organization. I have no doubt he will go far in his career — he's proven it time and again.

Irene Di Martino, PhD

CEO at AmpX

Technical leadership

Raises the technical bar and brings the team along.

I can always rely on Dean to go the extra mile. He consistently delivers more than expected, often studying new techniques to enhance his work. His ability to quickly transform from a developer to a data scientist through self-learning is remarkable. What truly sets Dean apart is his talent for explaining complex concepts to colleagues — his survey of time series methods became our team's foundational reference for all related projects.

Jan Bím, PhD

AI Department Lead, Datamole

Contact

Have a project in mind?

Let’s talk about how I can help bring your ideas to life with forecasting, optimisation, or automation that ships.

Common engagements

  • Forecasting systems
  • Pricing & optimisation
  • Operational automation
  • Advisory / fractional leadership

Want to sync live?

Schedule a quick call to scope work, timelines, and impact together.

Book a call