AI and R&D house

MicroscopeIT is something between a software house and a private research institute. About half of our employees have experience in Academia (and PhD titles to prove it), while the other half come from the industry. We are comfortable doing hard R&D projects, especially those concerning machine vision, machine learning, microscopic image analysis.

$50 - $99 / hr
10 - 49
Show all +
Wrocław, Poland
  • Kasztanowa 3a-5
    Wrocław, SL 53-125


Key clients: 

Aeolus Robotics


ClearLight Biotechnologies

Aquascope Project


Analytics Fire


Check our project with Netguru:


Sort by

Machine Learning Dev for Software Development Agency

“They were always responsive and attentive to our needs.”

Willing to refer: 
The Project
$50,000 to $199,999
Mar. - Nov. 2017
Project summary: 

MicroscopeIT guided a software development team and helped develop microservices and machine learning models for a third-party. They helped to do research and proposed solutions using the Python stack.

The Reviewer
51-200 Employees
Warsaw, Poland
Head of Sales, Software Company
The Review
Feedback summary: 

The work pleased the third-party recipient. The agency could not develop the solution without MicroscopeIT’s contributions. They supported the project with their strong expertise and understanding and integrated effortlessly into the agency’s structure.


Introduce your business and what you do there.

I’m a co-founder and the head of sales for a software development agency. I manage new business projects and development.


What challenge were you trying to address with MicroscopeIT?

We had a customer that required an advanced solution for which we didn’t have enough experience. They needed an advanced biometric and computer vision solution developed. We needed external help to kickstart the process, tell us which direction to go, and provide us with some fast results.


What was the scope of their involvement?

MicroscopeIT developers helped us with computer vision and machine learning components for a biometrics solution on behalf of our customer. MicroscopeIT used the same technology stack as us: Docker, TensorFlow, and other standard tools for the Python stack. They helped us develop microservices and machine learning models that we later used for the web and mobile applications developed by our team. MicroscopeIT worked on a specific part of the solution, mostly related to machine learning. They helped us come up with different models and train those models.

The requirements gathering was a collaborative effort. MicroscopeIT did a lot of research on their own, assessed the risks, and proposed multiple solutions to us. We decided together on which direction to take. This was not a standard Scrum project, but more of an R&D one.

What is the team composition?

There were around 15 developers from our side working on the backend, web, and mobile parts. We interacted with 3–5 developers from their team. Not all of them were engaged full-time. They worked in a model that consisted of 1–2 main developers who got advice from people with knowledge in the field.

How did you come to work with MicroscopeIT?

I met one of the owners of the company through the recommendation of a friend. They had very good credentials, having worked for Google in the past. They truly understood the subject matter.

How much have you invested with them?

The cost was around $50,000.

What is the status of this engagement?

We worked with MicroscopeIT from March– November 2017.


What evidence can you share that demonstrates the impact of the engagement?

We wouldn’t have been able to kickstart our own team without MicroscopeIT’s help. We were able to do this based on their initial input and help. The customer we developed the solution for is quite happy.

They were always responsive and attentive to our needs. On one occasion, we didn’t properly communicate the expected deadlines, but they delivered on time. We were very happy working with MicroscopeIT.

How did MicroscopeIT perform from a project management standpoint?

They worked closely with our team, which was good. They were self-organizing and plugged into our structure well. We all used Jira, Slack, Google Docs, and GitHub. Using the same set of tools was like working with a remote extension of our team. We’ve met their founders face-to-face, but we generally worked in a distributed manner.

What did you find most impressive about them?

We gave them a very difficult problem to solve, which we weren’t sure was actually solvable. MicroscopeIT was able to provide us with solutions and ideas on how to approach it that worked in the end. They truly understood the subject matter of biometrics, computer vision, and machine learning. This is a rare thing to find in a company.

Are there any areas they could improve?

They’re a small team, and we sometimes felt that we couldn’t get more help from them or that we couldn’t easily increase the team size. It wasn’t a big issue in the end, but I can imagine that it could be a problem for someone looking for a bigger team.

Do you have any advice for future clients of theirs?

It wasn’t a typical agile, end to end project so it was difficult to define its end goal or deliverables. It was a working solution, and we had to use certain parameters to define what success and failure meant. We could’ve found ourselves in the situation where, after a couple of months of development, the outcome was no longer doable. It was important to understand that we can’t always get a 100% outcome.

Overall Score
  • 5.0 Scheduling
  • 4.5 Cost
    Value / within estimates
  • 5.0 Quality
    Service & deliverables
  • 5.0 NPS
    Willing to refer