Research question
The question I am researching
From your experience, which creates more avoidable work?
- 01 Drawing / FAI preparation
- 02 Matching measurement results back to the compliance package
- ·· …or something else?
A one-line answer is genuinely useful. There is no wrong answer, and pointing me somewhere I have not considered ("or something else") helps just as much.
Research framing map
The inspection-documentation path I am studying
A common end-to-end path, shown to frame the research. This is not a claim about SpecTrace's scope, a product capability, or the only way teams work — it is the set of stages I am asking about.
- Drawing / revision received
- Ballooning & characteristic extraction
- Inspection planning
- CMM / measurement output
- Result-to-characteristic matching
- FAI / AS9102 / compliance package
// also relevant: GD&T interpretation · NCR / CAPA / SCAR · customer-portal uploads & revision changes
Who this is for
Perspectives especially useful
If your work touches any of these, your view would help. There is no obligation, and a referral to someone with a better vantage point is just as welcome.
- Quality leadershipQuality Manager, Director of Quality
- Metrology & CMMMetrology Manager, CMM Programmer
- Manufacturing & supplier qualityManufacturing Engineering Manager, Supplier Quality Manager
- AS9100 / AS9102 & metrology consultantsQuality and metrology consultants
Who is behind it
About the founder
Michael Pakyurek is a machine learning and computer vision engineer with experience building data-quality and applied vision systems. He is researching how quality and metrology teams handle inspection documentation through SpecTrace, an early research-stage project focused on understanding the workflow before building a product.
You can confirm who I am here: linkedin.com/in/michaelpakyurek, or email research@getspectrace.com.
Research boundaries
What this is, and what I will not ask you for
This is not a sales call and not a sales page. There is nothing to buy, and I am not claiming to have a finished tool, customers, or proof. I am simply looking to learn from people who do this work.
Learning about a workflow does not require sensitive material. A conversation or email exchange about SpecTrace does not require any confidential, ITAR, CUI, export-controlled, or customer-restricted documents, and I will not ask for them. I am interested in how the work flows and where it slows down, described in general terms.
Where this could go
A possible direction — only if the research justifies it
Nothing here is built. This is the direction SpecTrace could take only if the research points to a real, shared bottleneck — and it would be shaped with the people who helped find it. Conditional · not a roadmap commitment
- ListenUnderstand where inspection-documentation work creates avoidable effort, from people who do it. This is the current stage.
- FocusOnly if a sharp, shared pain emerges, scope one narrow workflow — not "everything."
- PrototypeA narrow, human-reviewed assistant concept for that one step; no autonomous quality decisions; tried only on non-sensitive or sample data with willing participants.
- Keep or dropKeep it only if it genuinely saves time and earns trust. If it does not, it does not get built.
// no timelines or commitments · human-reviewed, never autonomous · never any confidential, ITAR, CUI, export-controlled, or customer-restricted documents
Reply or refer
Share one workflow thought, or ask for more
The easiest way to help is one short email answer to the question above. If someone else would have a better view of this work, a referral is welcome. If an email exchange reaches a point where a brief call would be easier, we can arrange one by email — a call is optional and never required.