Skip to main content
Loading navigation
Back to challenges

Build a Gas Pipeline Tender Risk Agent with OpenAI Agents SDK & Galileo

Pipeline project bid delays create cost overruns due to complex procurement signals across North Africa. Build an autonomous agent using OpenAI Agents SDK and Galileo to evaluate tender delay risk and calculate project risk scores exceeding 90% accuracy.

Business OperationsHosted by Vera
Sign in free to start

What you are building

Utilize OpenAI Agents SDK with Galileo evaluation tracing to analyze cross-border pipeline procurement documents and generate risk mitigation schedules.

How work is evaluated

Evaluation

Evaluates the agent's ability to extract bid delay parameters and output formatted risk scores verified by Galileo metrics.

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

test_risk_score_range

Ensure calculated risk score falls between 0 and 1

Binary check

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

Dimension 2

extraction_precision

Precision of extracted tender parameters • target: 0.95 • range: 0.85-1

Binary check

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

Learning goals

  • Parse unstructured gas pipeline tender addendums and bilingual bid extensions

  • Implement autonomous risk scoring agents using OpenAI Agents SDK tool invocation

  • Evaluate prompt accuracy and response hallucinations using Galileo observability metrics

  • Produce automated schedule mitigation recommendations for energy megaprojects

Resources

Dataset notes

Synthetic dataset of Mediterranean gas pipeline tender notices, extension addendums, and historical concession agreement risks.

Challenge setup

Evaluates the agent's ability to extract bid delay parameters and output formatted risk scores verified by Galileo metrics.

Challenge input

JSON containing tender text and historical delay data

OpenAI Agents SDK

Official framework for multi-turn structured task execution

Galileo

Provides prompt and hallucination metrics

Evaluated output

JSON with delay_days, risk_score, and mitigation_steps

Evaluation checks
  • Ensure calculated risk score falls between 0 and 1
  • Precision of extracted tender parameters • target: 0.95 • range: 0.85-1
Configured evaluation evidence
  • Benchmark: ProcurementRiskBench-GCC
  • Extraction Precision target: 0.95
  • 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 · Optional
  • Bito AI · Optional
Observation
  • Galileo · Required
Reward / Eval
  • Galileo · Required

Evaluation contract

  • test_risk_score_range · Weight 1
  • extraction_precision · Weight 1

Metrics

  • extraction_precision · Target 0.95

Benchmarks

  • ProcurementRiskBench-GCC
  • energy

Reference cases

  • 1 public reference case

Workflow tasks

  1. Task 1

    Input: JSON containing tender text and historical delay data

    Output: JSON with delay_days, risk_score, and mitigation_steps

    Execution harness: Python

Evaluation tools

  • OpenAI Agents SDK · Agent Framework
  • Galileo · Evaluation Engine

Recipe freeze marker

No recipe freeze time is recorded.

Frequently Asked Questions about Build a Gas Pipeline Tender Risk Agent with OpenAI Agents SDK & Galileo