BEEBOP LABS
We are tackling the world's most complex power system challenges, assembling leading researchers and engineers to build the foundational AI that the energy transition depends on.
<5%
RMSE Day-ahead
98%+
Dispatch reliability
5,000+
CITATIONS
200K+
Assets CONNECTED
The Problem
The grid is becoming a decision problem.
Energy is the most important problem of our time. Food, water, security and computation all depend on solving it. The flexible capacity to solve it is already installed at the edge of the grid. Using it means making millions of decisions a second, under uncertainty, across markets, within physical limits.
Orchestration and decision-making at scale
Complexity grows super-linearly with fleet size but compute budgets do not. Scaling is the unsolved problem at the heart of the field, with implications stretching far beyond energy management alone.
Cross-asset, cross-market optimisation
Batteries, EVs and other DERs behave and respond differently, but they need to be detected, modelled, and aggregated into one compact representation that can deliver flexibility with bankable, infrastructural-grade, reliability. Optimally placing that capacity across markets through high-frequency trading compounds complexity.
01
Identify
Flex language • what sits behind each meter
02
Model
Digital twin • with every local constraint
03
Aggregate
Thousands of futures, one prism
04
Trade
Day-ahead, Intraday, Imbalance
asks
mid 92.4
Bids
Data-centre energy management
Grid-friendly scheduling of training and inference jobs that optimizes the grid's flexibility.
The research
Research threads
Our research starts with problems from the field, pursued together with an international network of leading institutes and academic labs. These are active research directions, some of which will reach the grid; all of them improve our understanding of the energy system.
Sequential decision-making under uncertainty
Trading and control where the world is stochastic and the model is not linear. Reinforcement learning, probabilistic forecasting and mathematical programming, combined with expert domain knowledge.
Agentic Energy Management
Using agents to plan and act across the energy stack.
Research infrastructure
Frontier research needs standard rails. Alongside the threads above, we build shared infrastructure the field is missing: connectivity abstraction across device makers, interfaces that couple assets to markets, and tools for tariff and market design. Engineering in service of the science.
THE RESULTS
Hard problems, worked to a result.
Each study follows the same arc: a problem without a known solution, the approach we took, and the result. Further studies will be added as work is published.
Aggregate and decompose
problem
A trading desk cannot optimise a million devices. Every asset added to a fleet multiplies the decision space; every OEM adds its own constraints, latencies and failure modes. The naive formulation is computationally intractable.
approach
We developed a representation that aggregates the flexibility of arbitrarily many heterogeneous assets into a few hundred parameters: energy constraints, state-dependent power limits, cost functions and forecasts, independent of fleet size. Trading stacks optimise against the compact model; schedules are then decomposed back into device-level actions that respect every local constraint, from user comfort to warranty limits.
result
One interface between any fleet and any trading stack, with dimensionality that stays flat as the fleet grows. The architecture is patented and is the subject of our world models essay.
THE RECORD
Selected publications
Research published by the people building Beebop.
The TEAM
Some of the people behind the models
This team has scaled flexibility platforms from zero to millions of assets, twice. Backgrounds span Tesla, Palantir, Centrica, Goldman Sachs and Bain.
Chief product officer
Bert Claessens
3,000+ citations in physics-informed AI for power systems. Fifteen years of published research, in production.
Head of Product
Evelyn Heylen
Research experience across low-voltage and high-voltage power systems. Industrial research with multiple European TSOs.
Head of Artificial Intelligence
Nikolaus Houben
PhD in machine learning for power systems, TU Wien and Berkeley Lab. Over 6 years of experience in research and industry.
10+
PhDs in AI, energy systems and optimisation
5,000+
Academic Citations
15+
Years Operating GW-Scale Flexibility
2
VPP Exits
#1
Ranked VPP Technology Globally by Guidehouse
4
Continents of GW-scale operations
Our team comes from world-leading research institutes and academic labs.
Work with us
The hard problems are still open.
Two ways to work with us.
Join the team
We hire researchers and engineers who want their work deployed on the real grid. There’s no application portal; write to us directly.
Research collaboration
For labs, institutes and industry teams working on adjacent problems.