Skip to main content
Loading navigation
Back to challenges

Middle East 2050 Grid Capacity Forecast Agent with OpenAI Agents SDK

Long-term grid planning across the Middle East requires modeling explosive power demand from sovereign AI datacenters and desalinization plants. Build an autonomous agent using OpenAI Agents SDK with function calling to execute load-forecasting simulations and recommend regional grid interconnect capacity, achieving sub-5% MAPE on 2030-2050 projection benchmarks.

Business OperationsHosted by Vera
Sign in free to start

What you are building

Construct an autonomous grid planning agent using OpenAI Agents SDK with function calling to forecast 2050 regional Middle East electricity demand.

How work is evaluated

Evaluation

Evaluates power demand forecast error (MAPE) and grid expansion recommendations.

Datasets

Public datasets, plus the private datasets you have access to.

How submissions are scored

Dimensions
2 scoring checks
Binary
2 pass or fail dimensions
Ordinal
0 scaled dimensions
Dimension 1

forecast_doubling_check

Must project demand doubling (multiplier >= 2.0) by 2050 in line with Rystad report

Binary check

This dimension contributes its full weight only when the submission satisfies the requirement. Partial credit is not awarded.

Dimension 2

mape_error_rate

Mean Absolute Percentage Error relative to baseline scenario benchmark • target: 4.5 • range: 0-20

Binary check

This dimension contributes its full weight only when the submission satisfies the requirement. Partial credit is not awarded.

Learning goals

  • Utilize OpenAI Agents SDK (`Runner`, `Agent`, `@function_tool`) for autonomous power simulation

  • Design load forecast simulation tools modeling AI datacenter growth and desalination growth

  • Formulate GCC Interconnection Grid transfer expansion recommendations

  • Evaluate load model accuracy using Mean Absolute Percentage Error (MAPE) targets

Resources

Dataset notes

Middle East 2025-2050 power demand projections, AI datacenter MW schedules, and Rystad Energy research data.

Challenge setup

Evaluates power demand forecast error (MAPE) and grid expansion recommendations.

Challenge input

JSON containing baseline load GW, AI load growth factor, and target forecast year.

OpenAI Agents SDK

Required OpenAI framework for autonomous function-calling agents

Evaluated output

JSON containing projected demand GW, required solar/nuclear mix %, and interconnect recommendation.

Evaluation checks
  • Must project demand doubling (multiplier >= 2.0) by 2050 in line with Rystad report
  • Mean Absolute Percentage Error relative to baseline scenario benchmark • target: 4.5 • range: 0-20
Configured evaluation evidence
  • Benchmark: Energy Grid Forecasting Benchmark
  • Mape Error Rate target: 4.5
  • 1 public reference case
Configured runtime
  • Python execution harness
  • Python sandbox (unavailable on Versalist)

Shows the first task, up to three tools, and up to three evidence items.

View technical recipe

Configured tools

Action Space
  • OpenAI · Required
  • Deepgram · Optional
  • Adept · Optional

Evaluation contract

  • forecast_doubling_check · Weight 1
  • mape_error_rate · Weight 1

Metrics

  • mape_error_rate · Target 4.5

Benchmarks

  • Energy Grid Forecasting Benchmark
  • energy

Reference cases

  • 1 public reference case

Workflow tasks

  1. Task 1

    Input: JSON containing baseline load GW, AI load growth factor, and target forecast year.

    Output: JSON containing projected demand GW, required solar/nuclear mix %, and interconnect recommendation.

    Execution harness: Python

Evaluation tools

  • OpenAI Agents SDK · Agent Framework

Recipe freeze marker

No recipe freeze time is recorded.

Frequently Asked Questions about Middle East 2050 Grid Capacity Forecast Agent with OpenAI Agents SDK