Free, no call
Send us 30 images. We will tell you if this works.
No meeting, no discovery call, no proposal. Sample images from the line and a description of what you need to catch, and you get a written memo back in a few days.
Vision Feasibility Check
The situation
Before we start
Every vendor wants a call first
You cannot get a straight technical answer without sitting through a pipeline stage. This skips it.
Accuracy claims with no basis
Numbers quoted before anyone has seen your parts are marketing. A band derived from your own images is an assessment.
The lighting problem nobody mentions
Most accuracy ceilings are set by optics and lighting, not the model. The memo says so when your images show it.
What working means here
Signed first. Measured after.
The value of each one is set with you during scoping, from your data and your line. What does not change is that they are written down first and published against afterwards, pass or fail.
- 01
Accuracy band
A realistic range for your classes, derived from your own images
- 02
Hardware class
What the frame rate and latency would actually require at the line
- 03
Go or no-go
A recommendation in writing, including fix-the-lighting-first when that is the answer
How it is assembled
Stage by stage
Each one ends in something you keep.
- Step 01
Send samples
Thirty to fifty images covering good parts, the defects you need to catch, and any awkward lighting conditions.
- Step 02
Describe the job
What has to be detected, how fast the line runs, and what should happen when something is found.
- Step 03
We review
An engineer looks at the images against the requirement and writes up what is achievable.
- Step 04
You get a memo
Accuracy band, hardware class, risks and a recommendation, in writing, within a few days.
Scope
What the work includes
- Review of 30 to 50 sample images from your process
- A realistic accuracy band for the classes you describe
- The hardware class the frame rate and latency would require
- The risks visible in your images, lighting and optics first
- A go, no-go, or fix-this-first recommendation
- Written and sent back. No call required at any point
Deliverables
What you keep
- 01A written feasibility memo, typically two to three pages
Manufacturing · Vision
Surface defect inspection, decided at the camera in 12 ms
Final inspection was manual and sampled, so surface defects were being found by the customer rather than on the floor. We surveyed the cell, captured production across three shifts, and commissioned an edge vision station that inspects every part and drives the existing reject gate.
Read the engagement- 12 ms
- decision latency at the camera
- 99.4%
- detection rate on hold-out set
- 100%
- of parts inspected, up from sampled
Start
How Vision Feasibility Check begins
Free written memo
Where an engagement lands inside its band is set by data readiness, integration depth, the reliability bar, evaluation burden, deployment constraints and usage economics. All six, explained.
No proposal, no pipeline stage. The next step is the engagement itself.
Send sample imagesFAQ
Asked before signing
Because it qualifies both of us faster than a sales call, and because a written technical answer is a better demonstration of how we work than a deck.
Reviewed by an engineer, held only as long as the assessment takes, then deleted. They are never used as training data. We will sign an NDA first if you prefer.
Yes, and that is the point. A memo that says the lighting sets a ceiling you cannot live with has saved you a project.
A few working days from receiving usable samples.
Often paired with
Where this fits with the rest
- 01Physical AI
Edge AI & Computer Vision
Detection, inspection, tracking and safety on-device. Milliseconds, and no network dependency.
- 02Start here
AI Pilot
Four weeks, fixed price, signed acceptance criteria. Working software, or a written case for stopping.
- 03Physical AI
Vision Systems Integration
Cameras, optics, lighting, enclosures, and the PLC, SCADA and MES interfaces that make results act.