Skip to main content
Back to challenges
Challenge

Build a Google ADK Streaming Ad Insertion & Monetization Agent

Media SaaS infrastructure platforms managing thousands of fast-channels require real-time ad slot decision engines. Build a Google ADK agent powered by Gemini models that evaluates viewer contextual signals and schedules target ad insertions with sub-200ms decision latency.

Business OperationsHosted 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 a Google ADK ad insertion agent that selects dynamic video ad placements based on viewer context and channel revenue goals.

Delivery guide

How work is evaluated

Evaluation

Evaluates ad placement selection and response execution speed.

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

latency_check

Verifies decision time is under 200ms

Binary check

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

Dimension 2

decision_time_ms

Execution latency for ad placement decisions • target: 140 • range: 0-1000

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

  • Build Google ADK dynamic tool components for ad inventory lookup

  • Process real-time video stream metadata and viewer demographic profiles

  • Optimize ad CPM fill rates while preserving viewer retention constraints

  • Achieve ad slot decision latency under 200ms

Resources and assets

Reference links and supporting material

Dataset notes

Synthetic streaming video ad break cues and advertiser catalog metadata.

Challenge setup

Evaluates ad placement selection and response execution speed.

Challenge input

JSON stream cue event payload

Google ADK

Agent development kit for real-time streaming tools

Evaluated output

JSON ad decision payload with ad_id, cpm_usd, and decision_time_ms

Evaluation checks
  • Verifies decision time is under 200ms
  • Execution latency for ad placement decisions • target: 140 • range: 0-1000
Configured evaluation evidence
  • Benchmark: Real-time Ad Recommendation Benchmark
  • Decision Time Ms target: 140
  • 1 public reference case
Configured runtime
  • Python execution harness
  • Python 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
  • Google ADK · Required
  • Google · Optional
  • Google Gemini · Optional

Evaluation contract

  • latency_check · Weight 1
  • decision_time_ms · Weight 1

Metrics

  • decision_time_ms · Target 140

Benchmarks

  • Real-time Ad Recommendation Benchmark
  • general

Reference cases

  • 1 public reference case

Workflow tasks

  1. Task 1

    Input: JSON stream cue event payload

    Output: JSON ad decision payload with ad_id, cpm_usd, and decision_time_ms

    Execution harness: Python

Evaluation tools

  • Google ADK · Agent Framework

Recipe freeze marker

No recipe freeze time is recorded.

Frequently Asked Questions about Build a Google ADK Streaming Ad Insertion & Monetization Agent