
Customer story · LiR Labs
LiR Labs automates biological wastewater treatment with AI that reads microscope images. Its models needed annotators who can read activated sludge. Pathwize found them.
- 100
- applications from one call
- 84
- usable profiles
- 22
- experts matched to the criteria
- 7
- Tier 1, ready to start

The challenge
A skill you will not find in a general labeling workforce.
The models are only as good as the people who outline and classify the structures in every microscope image. That needs biology knowledge you cannot teach in a one-page guideline. In-house, LiR Labs had a small team that could label, but not the capacity to scale, and hiring specialists one by one would have taken months. LiR Labs defined exactly what it needed:
- Practical experience with activated sludge microscopy
- Recognizing filamentous bacteria by their morphotype
- Estimating the filament index of a sample
- Telling bulking sludge, compact flocs and foam apart
- Experience with image segmentation
The approach
From one call to an anonymized shortlist.
- 01
Targeted call for experts
One call shown to members whose field fits, from water science and microbiology to environmental and process engineering, instead of a generic job ad.
- 02
First review
100 applications came in, 84 complete and relevant enough to assess.
- 03
Scoring against the criteria
Every profile was checked against LiR Labs' hard criteria (activated sludge microscopy, filament morphotypes, filament index, bulking versus foam) and preferences such as segmentation experience.
- 04
Tiering
The strongest matches went into Tier 1, solid profiles that need a short onboarding into Tier 2, together with their weekly availability.
- 05
Anonymized shortlist
LiR Labs reviewed anonymized profiles with its ML team before meeting anyone.
- 06
Practical screening
Before work starts, each candidate labels a few real images, so self-assessment is checked against actual skill.

“Really strong profiles. Pathwize's selection matches our criteria better than we had expected.”
Benedikt Huber, Co-Founder, LiR Labs
Where it stands
What happens next.
In a niche where most labeling vendors cannot staff a single qualified annotator, the pool for the pilot stood within weeks. The annotation itself starts once LiR Labs' images and labeling guide are ready for external work, with every expert passing a test on real images first.
This story covers expert sourcing and screening. We will add results on data quality once the annotation has run.
Need expertsnobody else can find?
Tell us the profile. We search our network by field and test every candidate on your data.
Talk to us