Skip to main content
Loading navigation
Back to challenges

LlamaIndex Aeration Telemetry Diagnostic Agent with Gemini 2.5 Flash

Wastewater aeration retrofits often generate unstructured alarm logs and OPC UA telemetry spikes that overwhelm operators. Build an intelligent telemetry diagnostic agent using LlamaIndex and Gemini 2.5 Flash to automatically parse telemetry streams and detect physical failure states. Success is measured by achieving over 90% diagnostic accuracy on benchmarked aeration fault logs.

EngineeringHosted by Vera
Sign in free to start

What you are building

Implement a LlamaIndex RAG agent integrated with Gemini 2.5 Flash to parse real-time OPC UA aeration telemetry and identify physical failure states.

How work is evaluated

Evaluation

Evaluates diagnostic accuracy and root-cause identification precision for aeration telemetry anomaly logs.

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

valid_json_structure

Ensures output includes valid fault_code, root_cause, and confidence_score.

Binary check

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

Dimension 2

diagnostic_accuracy

Percentage of correctly identified fault codes. • target: 0.9 • range: 0-1

Binary check

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

Learning goals

  • Ingest and index OPC UA time-series telemetry and system logs using LlamaIndex.

  • Integrate Gemini 2.5 Flash for high-speed anomaly diagnosis and root cause analysis.

  • Build a structured query agent capable of correlating blower pressure drops with dissolved oxygen anomalies.

  • Evaluate diagnostic accuracy against known fault ground-truth data.

Resources

Dataset notes

Time-series OPC UA telemetry stream from a municipal wastewater aeration retrofit, featuring pressure sensor, dissolved oxygen, and alarm event logs.

Challenge setup

Evaluates diagnostic accuracy and root-cause identification precision for aeration telemetry anomaly logs.

Challenge input

JSON object with telemetry_window array and alarm_events

LlamaIndex

Framework required for RAG and structured indexing over time-series logs.

Gemini 2.5 Flash

High speed and large context window suited for raw sensor log analysis.

Evaluated output

JSON object with fault_code, root_cause, and confidence_score

Evaluation checks
  • Ensures output includes valid fault_code, root_cause, and confidence_score.
  • Percentage of correctly identified fault codes. • target: 0.9 • range: 0-1
Configured evaluation evidence
  • Benchmark: IndustrialTelemetryEval-2026
  • Diagnostic Accuracy target: 0.9
  • 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
  • Llama Index · Required
  • LlamaIndex · Optional
Policy Serving
  • Gemini 2.5 Flash · Optional

Evaluation contract

  • valid_json_structure · Weight 1
  • diagnostic_accuracy · Weight 1

Metrics

  • diagnostic_accuracy · Target 0.9

Benchmarks

  • IndustrialTelemetryEval-2026
  • energy

Reference cases

  • 1 public reference case

Workflow tasks

  1. Task 1

    Input: JSON object with telemetry_window array and alarm_events

    Output: JSON object with fault_code, root_cause, and confidence_score

    Execution harness: Python

Evaluation tools

  • LlamaIndex · Agent Framework
  • Gemini 2.5 Flash · LLM Inference

Recipe freeze marker

No recipe freeze time is recorded.

Frequently Asked Questions about LlamaIndex Aeration Telemetry Diagnostic Agent with Gemini 2.5 Flash