Skip to main content
Back to challenges
Challenge

AutoGen High-Throughput Microdrama Streaming Agent on Vast.ai Runtime

Short-form video platforms experience immense streaming demand and content moderation challenges under new digital safety laws. Build a multi-agent moderation and transcoding dispatch pipeline using AutoGen and GPU compute provisioned on Vast.ai.

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.

Construct an automated video processing multi-agent swarm using AutoGen hosted on Vast.ai GPU infrastructure.

Delivery guide

How work is evaluated

Evaluation

Evaluates video frame safety parsing precision and cloud GPU task orchestration.

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

approval_status_valid

Verifies that clean videos are marked as APPROVED.

Binary check

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

Dimension 2

transcoding_speed

FPS processing throughput ratio • target: 0.98 • 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

  • Set up AutoGen conversational multi-agent swarm for video pipeline tasks

  • Deploy inference and video transcoding routines on Vast.ai instance runtimes

  • Verify compliance against newly introduced Data Centre & Content Safety regulations

Resources and assets

Reference links and supporting material

Dataset notes

Sample video meta logs with compliance violation annotations.

Challenge setup

Evaluates video frame safety parsing precision and cloud GPU task orchestration.

Challenge input

JSON video metadata and frame features

AutoGen

Mandatory framework requirement

Vast.ai

Assigned compute infrastructure

Evaluated output

JSON moderation clearance report

Evaluation checks
  • Verifies that clean videos are marked as APPROVED.
  • FPS processing throughput ratio • target: 0.98 • range: 0-1
Configured evaluation evidence
  • Benchmark: Video-Mod-Bench
  • Transcoding Speed target: 0.98
  • 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
  • AutoGen · Required
  • Vast.ai · Optional
  • AWS · Optional

Evaluation contract

  • approval_status_valid · Weight 1
  • transcoding_speed · Weight 1

Metrics

  • transcoding_speed · Target 0.98

Benchmarks

  • Video-Mod-Bench
  • security

Reference cases

  • 1 public reference case

Workflow tasks

  1. Task 1

    Input: JSON video metadata and frame features

    Output: JSON moderation clearance report

    Execution harness: Python

Evaluation tools

  • AutoGen · Multi-Agent Framework
  • Vast.ai · Compute Cloud Engine

Recipe freeze marker

No recipe freeze time is recorded.

Frequently Asked Questions about AutoGen High-Throughput Microdrama Streaming Agent on Vast.ai Runtime