Multi-Agent BESS Capacity Auction Strategy with AutoGen and Google Jules
This challenge focuses on building a multi-agent simulation to optimize bidding strategies for Battery Energy Storage Systems (BESS) in the newly established German capacity market. Using Microsoft's AutoGen framework, you will design a system of collaborating agents: a 'Market Analyst' to forecast demand and prices, a 'BESS Operator' to calculate state-of-health and discharge costs, and a 'Bid Strategist' to synthesize these inputs into a winning auction bid. The implementation must incorporate Google Jules for AI-assisted engineering and code generation of the underlying physics-based battery models. Participants will tackle the complexities of energy arbitrage and capacity availability requirements. The goal is to maximize revenue while adhering to strict market regulations (such as those recently announced in Germany and Poland). You will need to implement complex logic for state-of-charge (SOC) management and cycling constraints, ensuring that the agents communicate effectively to reach a consensus on the optimal bid price and volume for a 200MW/800MWh BESS facility.
What you are building
The core problem, expected build, and operating context for this challenge.
This challenge focuses on building a multi-agent simulation to optimize bidding strategies for Battery Energy Storage Systems (BESS) in the newly established German capacity market. Using Microsoft's AutoGen framework, you will design a system of collaborating agents: a 'Market Analyst' to forecast demand and prices, a 'BESS Operator' to calculate state-of-health and discharge costs, and a 'Bid Strategist' to synthesize these inputs into a winning auction bid. The implementation must incorporate Google Jules for AI-assisted engineering and code generation of the underlying physics-based battery models. Participants will tackle the complexities of energy arbitrage and capacity availability requirements. The goal is to maximize revenue while adhering to strict market regulations (such as those recently announced in Germany and Poland). You will need to implement complex logic for state-of-charge (SOC) management and cycling constraints, ensuring that the agents communicate effectively to reach a consensus on the optimal bid price and volume for a 200MW/800MWh BESS facility.
Shared data for this challenge
Review public datasets and any private uploads tied to your build.
How submissions are scored
These dimensions define what the evaluator checks and which criteria separate a passable run from a strong one.
SOC Constraint Test
Ensures the battery never discharges below 10% SOC based on the bidding strategy.
This dimension contributes its full weight only when the submission satisfies the requirement. Partial credit is not awarded.
Revenue Efficiency
Ratio of achieved revenue to theoretical maximum revenue. • target: 0.85 • range: 0-1
This dimension contributes its full weight only when the submission satisfies the requirement. Partial credit is not awarded.
What you should walk away with
Implement AutoGen group chats for collaborative energy market strategy development
Master prompt engineering for specific energy domain roles using Google Jules as a coding assistant
Design complex reward functions for energy arbitrage that account for battery cycling degradation
Orchestrate human-in-the-loop validation for high-stakes financial bidding simulations
Deploy localized LLMs to handle sensitive grid telemetry and market pricing data
Optimize agent communication frequency to minimize token usage while maintaining strategy coherence
Validate bidding strategies against historical German frequency containment reserve (FCR) data
How this agent runs
The system will be evaluated on its ability to generate bids that maximize revenue without violating battery health constraints over a 30-day simulated period.
Challenge input
JSON containing historical market prices and battery state
AutoGen
Enables multi-agent collaboration for complex bidding strategies.
Google Jules
Assists in coding the underlying physics and market models.
Evaluated output
JSON bid object
- Ensures the battery never discharges below 10% SOC based on the bidding strategy.
- Ratio of achieved revenue to theoretical maximum revenue. • target: 0.85 • range: 0-1
- Benchmark: EnergyArbitrageBench
- Revenue Efficiency target: 0.85
- 1 public reference case
- Python execution harness
- Python sandbox
View technical recipe
Configured tools
- Google · Required
- Google Jules · Optional
- AutoGen · Optional
- Google Jules · Optional
Evaluation contract
- SOC Constraint Test · Weight 1
- Revenue Efficiency · Weight 1
Recipe state
This is a preview. The configuration can change before the evaluation recipe is locked.
Run this agent on your dataset and AI stack
Bring your dataset, model providers, and success criteria. We will scope the right managed run for your team.
Scope a managed run[ok] Wrote CHALLENGE.md
[ok] Wrote .versalist.json
[ok] Wrote eval/examples.json
Requires VERSALIST_API_KEY. Works with any MCP-aware editor.
DocsFind another challenge
Jump to a random challenge when you want a fresh benchmark or a different problem space.