# Korbinian Kuusisto

Korbinian Kuusisto makes videos testing how iPhones and machine-vision sensors can inspect products on manufacturing lines, including detecting defects and checking labels. Kuusisto also writes about Berlin’s startup scene and local politics.

- Profile: https://b2binfluencer.marketing/creators/korbinian-kuusisto/
- LinkedIn: https://www.linkedin.com/in/korbinian-kuusisto-58a6b170
- LinkedIn followers: 5,981
- Median of last 5 posts: 29 reactions, 1 comments (checked 2026-11-02)
- Topics: startup, industrial-automation
- Based on 9 dated posts, latest 2026-09-30

## Summary

Korbinian Kuusisto is most useful for filmed, hands-on tests of iPhones and machine-vision sensors on manufacturing lines.

- Best for: Machine-vision or industrial-automation brands wanting practical plant-floor video for quality and process engineering audiences.
- Not for: Campaigns needing guaranteed reach into defined manufacturing buyer slices or fully approval-led scripted product messaging.
- Check first: Confirm recent views, watch-through, audience job function, employer type, seniority, and geography.

### Takeaways

- An iPhone on a ceramic tile line detected a defect that triggered a physical reject.
- An iPhone and a sensor were trained on three bottles and tested on two unseen.
- After a sensor correction, Kuusisto retested label presence and alignment on Fritz-Limo bottles.

## Main ideas

- **A phone can now do work that used to require industrial vision hardware**: Kuusisto's central claim is that a consumer device has crossed into a job purpose-built industrial cameras have done for decades, and that this changes what a factory can install and at what cost. The argument is made by showing an installation rather than describing the idea, so the reader sees the inspection decision and the physical action that follows it.
- **Industry claims about 'AI smart sensors' should be tested, not repeated**: Kuusisto treats the machine-vision sector's marketing as a claim to be checked. The posts set up head-to-head tests against the hardware the industry sells for modern AI manufacturing and state the doubt openly before showing the result, which puts the reader in the position of judging the claim rather than receiving a specification.
- **A vendor correction is a reason to test more precisely, not to abandon the comparison**: When a manufacturer disputes what a test actually measured, Kuusisto accepts the point and rebuilds the test around the capability the vendor documents. This keeps the original result standing while narrowing the question, and it credits the correction on camera instead of quietly dropping the earlier video.
- **Start with something ordinary so a technical test stays legible**: Several tests begin with an object or setting a non-specialist already understands, which gives a general audience a way into a measurement question. The everyday framing is not decoration: it sets up the actual capability being tested, such as how small a defect a consumer device can resolve.
- **Berlin's startup scene should own its part in the city's cost problem**: Written away from manufacturing, this argument says the startup scene spent years treating rising rents as a neutral market outcome while also warning the city that automation will take its jobs, and then reacted with outrage when voters turned against it. The political position is stated plainly, including where Kuusisto disagrees with the parties involved, and it sits on the same account as the industrial testing posts.

## Work to read

- [Installing an iPhone on a ceramic tile production line](https://www.linkedin.com/posts/korbinian-kuusisto-58a6b170_manufacturing-qualitycontrol-industrialautomation-activity-7498015125136822272-sHop)
- [Testing an AI smart sensor against an iPhone](https://www.linkedin.com/posts/korbinian-kuusisto-58a6b170_machinevision-defectdetection-injectionmolding-activity-7505171305890070528-K-lW)
- [Rechecking Fritz-Limo labels after a sensor correction](https://www.linkedin.com/posts/korbinian-kuusisto-58a6b170_machinevision-labelinspection-manufacturing-activity-7510980141041422337-DDWr)
- [Inspecting a single grain of sugar with an iPhone](https://www.linkedin.com/posts/korbinian-kuusisto-58a6b170_how-small-a-defect-can-an-iphone-detect-activity-7488962572986142720-kWJC)
- [On Berlin's election and the startup scene's reaction](https://www.linkedin.com/posts/korbinian-kuusisto-58a6b170_berlin-startup-elections-activity-7508482692523466752-t0Eg)

## Brand relationships

- **KEYENCE Corporation** (mention)
- **fritz-kola GmbH** (mention)

## Partnerships

No sponsor policy, rate card, media kit, or booking route appears in the material reviewed, so this section does not describe a published partnership product for Korbinian Kuusisto. What the posts do document is a recurring pattern of hands-on hardware testing filmed on live production lines: an iPhone used to catch defects at a ceramic tile factory in eastern Germany, a machine-vision smart sensor tested against an iPhone on label presence and alignment, and an AI smart sensor run against an iPhone in an injection-moulding defect check. The label test followed direct feedback from the sensor maker's team, which said the earlier test used the device outside its intended purpose; the follow-up video was rebuilt around that correction. That is evidence of technical exchange with a vendor, not evidence of payment or of an agreed commercial format. No deliverable set, exclusivity, usage rights, disclosure wording, turnaround time, or fee is stated anywhere in this material. A buyer should treat the format as open and negotiate production requirements directly: factory or line access, permission to film in a working plant, the specific device under test, and who supplies it. Because no partnership terms are documented either way, this is not a claim that Kuusisto never accepts sponsorship — only that the reviewed posts show none.

## Alternatives

- Chris Turlica: Choose Chris Turlica when the campaign decision sits above the line, with plant directors, operations leaders or manufacturing buying committees rather than the quality engineers who approve a vision system. The material is executive argument about manufacturing reliability: posts on the cost of unplanned downtime, on US factories running at 70–80% of capacity against reliability constraints, and on embedding AI in frontline workflows, with longer versions placed in business press such as Fast Company and Forbes. For a machine-vision or automation brand that wants to be read as a strategic authority on factory performance, and is comfortable selling through framing rather than through a demonstration, Turlica reaches the people who set reliability priorities and sign off budgets.
- Mutembei Kariuki: Choose Mutembei Kariuki when the brand's category is the broader physical-AI and industrial-robotics conversation rather than inspection hardware, and when the audience the campaign needs is the ecosystem around it: AI and robotics practitioners, integrators, and the investors and event network that shapes which approaches get attention. The posts work at that level, covering a Bay Area physical-AI event with live robotics and edge-AI demonstrations, and a reflection on Jidoka, Toyoda's principle of machines stopping themselves when they detect an abnormality, and what it means for machines that can now detect defects themselves. For a brand selling edge AI, orchestration or robotics into manufacturing, that ecosystem positioning is closer to the buying conversation than a defect-inspection test.
- Holly Hunter: Choose Holly Hunter when the real decision is about B2B audience and channel strategy rather than manufacturing quality, when a brand selling machine vision or automation wants its marketing team to reach B2B marketing and brand decision-makers, or wants content about where professional audiences now spend attention. The posts are built for that reader: YouTube overtaking ITV in UK home viewing and what it means for B2B brands, the blurring line between consumer and professional media habits, and a breakfast event on B2B marketing trends. This is the closer fit only when quality engineering is not the target and channel advice to a marketing-industry audience is.

---
AI-assisted summary of public posts. Audience demographics are not available. Engagement appears only where we checked recent posts. Method: https://b2binfluencer.marketing/methodology/