Skip to main content
Loading navigation
Back to challenges

Software-Defined Control Logic Engine with Vercel AI SDK

Industrial control logic is rapidly transitioning from dedicated hardware PLCs to cloud-connected software-defined control containers. You will build an edge-to-cloud distributed control logic orchestrator using Vercel AI SDK to stream real-time operational adjustments directly to edge execution runtimes. The pipeline must support continuous stream processing with zero stream dropouts and sub-100ms response delivery.

EngineeringHosted by Vera
Sign in free to start

What you are building

Develop a TypeScript control system using Vercel AI SDK streaming capabilities to deliver low-latency dynamic control directives to soft-PLCs.

How work is evaluated

Evaluation

Evaluates streaming validity and response time of generated software-defined control directives.

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

stream_completion_check

Validates that stream closed successfully with valid JSON EOF marker.

Binary check

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

Dimension 2

first_chunk_latency_ms

Time to receive first control streaming chunk • target: 80 • range: 0-300

Binary check

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

Learning goals

  • Utilize Vercel AI SDK `streamText` and `generateObject` primitives for streaming soft-PLC control directives

  • Implement edge state updates using TypeScript async generators

  • Construct soft-PLC execution adapters consuming JSON control streams

  • Evaluate streaming throughput and chunk latency on simulated industrial edge nodes

Resources

Dataset notes

Continuous process telemetry records from high-speed bottling and packaging machinery.

Challenge setup

Evaluates streaming validity and response time of generated software-defined control directives.

Challenge input

JSON specifying line_demand_units_sec and current_belt_speed_m_s

AI SDK

Required framework assignment.

Evaluated output

Streamed JSON chunks culminating in target_belt_speed_m_s and motor_voltage_v

Evaluation checks
  • Validates that stream closed successfully with valid JSON EOF marker.
  • Time to receive first control streaming chunk • target: 80 • range: 0-300
Configured evaluation evidence
  • Benchmark: Software-Defined Control Streaming Benchmark
  • First Chunk Latency Ms target: 80
  • 1 public reference case
Configured runtime
  • JavaScript execution harness
  • javascript 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
  • Vercel AI SDK · Required
Orchestration
  • Vercel AI SDK · Required

Evaluation contract

  • stream_completion_check · Weight 1
  • first_chunk_latency_ms · Weight 1

Metrics

  • first_chunk_latency_ms · Target 80

Benchmarks

  • Software-Defined Control Streaming Benchmark
  • general

Reference cases

  • 1 public reference case

Workflow tasks

  1. Task 1

    Input: JSON specifying line_demand_units_sec and current_belt_speed_m_s

    Output: Streamed JSON chunks culminating in target_belt_speed_m_s and motor_voltage_v

    Execution harness: JavaScript

Evaluation tools

  • AI SDK · Agent & Streaming SDK

Recipe freeze marker

No recipe freeze time is recorded.

Frequently Asked Questions about Software-Defined Control Logic Engine with Vercel AI SDK