You will get Industrial Computer Vision Paid PoC - Defect Detection
Project details
A 3-6 week paid PoC that verifies the feasibility of surface defect detection, object detection, or action recognition after samples, labels, test conditions, and acceptance metrics are agreed. Data quality assessment, problem definition, baseline model, offline evaluation, result report, and validation procedure. An edge-deployment demo is included only if defined in the contract. Production deployment, optical hardware, on-site implementation, and maintenance are quoted separately.
What's included
| Service Tiers |
Starter
$5,000
|
Standard
$10,000
|
Advanced
$15,000
|
|---|---|---|---|
| Delivery Time | 21 days | 30 days | 42 days |
Number of Revisions | 0 | 0 | 0 |
Model Validation/Testing | - | - | - |
Model Documentation | - | - | - |
Data Source Connectivity | - | - | - |
Source Code | - | - | - |
About Zhang
Enterprise AI Solutions Lead | RAG, Document Intelligence & Workflows
Beijing, China - 7:47 am local time
I lead enterprise software and AI solution delivery at Beijing Hengai Technology Development Co., Ltd. Founded in 2006, the company has delivered more than 300 projects. Its National High-Tech Enterprise status is valid through 2028. We deliver enterprise AI services under the Lingxi Link Digital Employee brand.
Core services:
- Enterprise RAG systems: document parsing, metadata, hybrid retrieval, reranking, source citations, permission-aware access, version traceability, and evaluation
- Document intelligence: OCR and parsing, field extraction, confidence scoring, validation rules, exception queues, human review, change tracking, and structured export
- Human-in-the-loop AI workflows: controlled tool use, input checks, approval points, failure handoff, execution logs, and auditable outputs
Multi-tenant architecture, Java/Spring Boot integration, private deployment, and model and knowledge governance provide the delivery foundation. Industrial vision is handled separately as a paid PoC when representative samples, test conditions, and acceptance metrics are available.
My preferred engagement starts with a focused paid discovery or pilot. Before production implementation, we confirm the data, permissions, interfaces, expected outputs, evaluation set, human review points, milestones, and operational risks.
I do not promise guaranteed accuracy, fully autonomous operation, or a fixed production estimate before the evidence and system constraints are reviewed. The first milestone should solve one clear workflow and produce written acceptance results.
Steps for completing your project
After purchasing the project, send requirements so Zhang can start the project.
Delivery time starts when Zhang receives requirements from you.
Zhang works on your project following the steps below.
Revisions may occur after the delivery date.
Data assessment and problem definition
Assess sample quality and define the defect or action classes, test conditions, and acceptance metrics
Baseline model and offline evaluation
Build a baseline model and run offline evaluation against the agreed frozen test set