Skip to main content
Back to challenges
Challenge

Build a Wastewater Telemetry Anomaly Detector with AI SDK and o3

Aging wastewater aeration control systems suffer from unexpected dissolved oxygen drops and unrecorded PLC sensor drifts. Build a real-time SCADA telemetry analysis pipeline using AI SDK and o3 reasoning model to classify aeration anomalies and achieve over 95% detection accuracy.

EngineeringHosted by Vera
Sign in free to start
Challenge brief

What you are building

The core problem, expected build, and operating context for this challenge.

Develop an automated telemetry analyzer for PLC aeration retrofits using AI SDK tools paired with o3 model reasoning.

Delivery guide

How work is evaluated

Evaluation

Evaluates the accuracy and response time of the AI SDK telemetry classifier on simulated wastewater PLC streams.

Datasets

Challenge datasets

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

Evaluation rubric

How submissions are scored

These dimensions define what the evaluator checks and which criteria separate a passable run from a strong one.

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

valid_json_structure

Checks if output contains anomaly_detected and root_cause fields.

Binary check

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

Dimension 2

f1_score

F1 score of anomaly detection on test telemetry stream • target: 0.95 • 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

What you should walk away with

  • Stream OPC UA telemetry into AI SDK structured tool invocations

  • Leverage OpenAI o3 reasoning to diagnose dissolved oxygen anomalies

  • Build real-time alert triggers for PLC retrofits with zero false positives

  • Evaluate operational reliability metrics against baseline historical logs

Resources and assets

Reference links and supporting material

Dataset notes

Simulated time-series telemetry from a 4-basin municipal wastewater aeration facility.

Challenge setup

Evaluates the accuracy and response time of the AI SDK telemetry classifier on simulated wastewater PLC streams.

Challenge input

JSON object with time-series sensor array (do_level, airflow_rate, power_kw, timestamp)

AI SDK

Required framework for AI-powered TS applications

o3

Deep reasoning model for physical telemetry root-cause diagnosis

Evaluated output

JSON object with anomaly_detected (boolean), root_cause (string), confidence (float)

Evaluation checks
  • Checks if output contains anomaly_detected and root_cause fields.
  • F1 score of anomaly detection on test telemetry stream • target: 0.95 • range: 0-1
Configured evaluation evidence
  • Benchmark: SCADA-Bench-2025
  • F1 Score target: 0.95
  • 1 public reference case
Configured runtime
  • JavaScript execution harness
  • javascript sandbox (unavailable on Versalist)

Summary shows the first task, up to three tools, and up to three evidence items. Open the technical recipe to view all configured items.

View technical recipe

Configured tools

Action Space
  • Vercel AI SDK · Optional
  • OpenAI o3 · Optional
Policy Serving
  • o3 · Required
Orchestration
  • Vercel AI SDK · Optional

Evaluation contract

  • valid_json_structure · Weight 1
  • f1_score · Weight 1

Metrics

  • f1_score · Target 0.95

Benchmarks

  • SCADA-Bench-2025
  • energy

Reference cases

  • 1 public reference case

Workflow tasks

  1. Task 1

    Input: JSON object with time-series sensor array (do_level, airflow_rate, power_kw, timestamp)

    Output: JSON object with anomaly_detected (boolean), root_cause (string), confidence (float)

    Execution harness: JavaScript

Evaluation tools

  • AI SDK · agent_framework
  • o3 · llm_model

Recipe freeze marker

No recipe freeze time is recorded.

Frequently Asked Questions about Build a Wastewater Telemetry Anomaly Detector with AI SDK and o3