Data Science as a Service

MindCraft is a leader in Machine Learning and Artificial Intelligence development and consulting.  
Proven Data Science professionals at MindCraft have already delivered over 15 projects of various complexity for startups and enterprise-level clients, including Fortune 500 companies.

Our Specialties:

  • Machine Learning
  • Artificial Intelligence
  • Computer Vision
  • Natural Language Processing
  • Business Intelligence

Our Services:

  • Data Science R&D
  • AI Proof of Concept
  • ML Model Development and Integration
  • AI & ML Consulting

Why MindCraft:

  • Certified Data Science professionals
  • Ph.D. degree holders in Engineering and Applied Maths
  • 20+ years of software development experience
  • Solid Business Analysis expertise
  • Industry and community leadership, participation in international events
  • Mentorship at the biggest DS conference in Ukraine
  • Proven track record across domains: Banking, Retail, Real Estate, e-Commerce, Marketing, Hospitality
  • Guidance and support throughout the whole software development lifecycle, extended business value
 
$1,000+
 
$50 - $99 / hr
 
10 - 49
 Founded
2017
Show all +
Lviv, Ukraine
headquarters

Portfolio

Key clients: 

Finance & Banking, Marketing & Advertising, E-commerce, Retail, Manufacturing, Entertainment & Hospitality, Automotive. 

 

PoC version, our focus was to improve the speed of photo processing and the speed of learning new styles.

Already on the first model, the engine showed the speed of photo processing less than one second. As for the basic quality, you can view the photos on our site. 

  • Speed of photo processing less than one second
  • Ability to quickly learn new styles

If you are looking to build an AI

& ML solution like this, contact us [email protected]
Predictive Sales Analytics Tool for Special Offers Evaluation Image

Predictive Sales Analytics Tool for Special Offers Evaluation

For a dealer in the car parts retail business, a team of machine learning developers at MindCraft created a predictive sales analytics solution.

  • It automatically measures the profitability of buying items from the special offers lists
  • The solution evaluates the possible sales rates and estimates the discount
  • It provides invaluable marketing insights and helps to make smart business
decisions.

In fact, the predictive sales analytics method we used can be of a great benefit to any B2B or B2C retail business. With a little customization, this data-driven technology can help organizations quickly make the right marketing decisions.

f you are looking to build an AI & ML solution like this, contact us [email protected]

Machine Learning-Based Sales Forecasting Tool for Automotive Image

Machine Learning-Based Sales Forecasting Tool for Automotive

A Machine Learning system we developed for the customer can help achieve a significant Inventory Turnover Ratio and thus increase revenue per dollar invested in stock items

Benefits you'll get with this Engine:

  • This approach applies to any retail business.
  • Predict sales of any product in stock and boost the general Business Intelligence.
  • The Inventory Turnover Ratio of 80% in two
weeks. 

If you are looking to build an AI & ML solution like this, contact us [email protected]

Automated Document Classification Solution For Banking Image

Automated Document Classification Solution For Banking

Model is ready to be put into production and used for document classification, we received 99% of accuracy on the training data and around 90% on the test datasets

Benefits you'll get with this Engine:

  • Classifying dozens of different document categories.
  • Classification and recognition of handwritten documents and mixed-type document
  • A system could automate various document processing
tasks which are now performed manually

If you are looking to build an AI & ML solution like this, contact us [email protected]

Technologies used:

Subject: Automated Document Classification with Machine Learning, Data extraction
Data Science areas: Natural Language Processing, Computer vision, Optical Character Recognition
Architectures: Logistic Regression, Random Forests
Tools: Python, Tensorflow, Sklearn, Tesseract

E-commerce Price Prediction System Image

E-commerce Price Prediction System

We managed to create a business intelligence system that not only predicts the prices of products but also measures its own accuracy

Benefits you'll get with this Algorithm: 

  • Almost 95% of average relative accuracy for product pricing.
  • Keep the error rate to the minimum
  • Opportunity use engine in other domains, Retail, E-commerce, hospitality
  • Individual engine development under
the business client.

If you are looking to build an AI & ML solution like this, contact us [email protected]

Technologies used:

Subject: Price Prediction Software, Prediction Accuracy Evaluation, Deep Learning
Data Science areas: Feature Analysis, Estimators, Machine Learning
Architectures: Linear Regression, Boosting Regressors, Deep Neural Networks
Tools: Python, Tensorflow, Sklearn

Reviews

Sort by

AI Architecture & Development for Sensitive Data Detection

"...they offered and over-delivered."

Quality: 
5.0
Schedule: 
5.0
Cost: 
5.0
Willing to refer: 
5.0
The Project
 
$10,000 to $49,999
 
Aug. - Nov. 2019
Project summary: 

MindCraft provided a dedicated team to assist with architecting and developing an AI solution for a third-party client. They worked in parallel with the in-house technicians, following two-week sprints.

The Reviewer
 
11-50 Employees
 
Zürich, Switzerland
Harro M. Wiersma
Founder & CEO, regarding data GmbH
 
Verified
The Review
Feedback summary: 

MindCraft came up with alternative approaches that were effective in developing the solution. Based on their work, the end client is now discussing future capabilities for a second version of the product. MindCraft demonstrated creative problem-solving skills and had a knack for communication.

The client submitted this review online.

BACKGROUND

Please describe your company and your position there.

As a Management Consultant Company with the main focus on Data and Analytics we support our customers around the topics of new disruptive technologies, data, digitalization and thought management.

I founded re:data in 2015 in Switzerland to help customers smartly use data that is in need of analysis and combine it with the customers desire to make an impact with it's business.

By smartly combining data we help our customers to understand trends and ideas that can be found in their data. This enables them to find the added value urgently needed.

OPPORTUNITY / CHALLENGE

For what projects/services did your company hire MindCraft?

We needed to have a dedicated team to extend our Artificial Intelligence capabilities with both the overall architecture definition and development. Our customer had an urgent regulatory compliance issue and in the pre-project discussions we agreed on a possible and suitable roadmap to tackle this.

What were your goals for this project?

integration of a Convolutional Neural Network with a user friendly frontend (developed by another dedicated service provider) to detect sensitive data in a wide range of files and services.

SOLUTION

How did you select this vendor?

we specifically were looking for a specialised AI provider in western Ukraine to build up a strong and long term partnership.

Before committing we have met in person with the Mindcraft Team and have commonly build up the strategy how to get the team involved in the best way possible.

After having spent in workshop style several parts of days with the team it was clear that the way of thinking, working and partnership was exactly what we were looking for.

Describe the project in detail.

the project (MVP) consisted of 3 months - divided into 2 weeks sprints.

As the cooperation on our side was based on 3 working parts (our local team, Mindcraft and a 3rd provider responsible for the frontend and integration) we decided initially to work in parallel and have team specific sprint- and status-meetings.

As soon as the components came close to full integration we decided to combine the meetings and integrate the development efforts.

Mindcraft has shown their brilliant capabilities and knowhow - even in difficult circumstances of development - as access to the end-customer was limited. They came up with creative but effective solutions and approaches to reach the project goals.

What was the team composition?

our local team in Switzerland consisted of a Project Manager, a Business Analyst (and Ukrainian Language Liaison for possible communication challenges due to the different native languages), an AI specialist and an technical architect.

The Mindcraft team consisted of 4 AI specialists and a Project Officer. For the front-end development we had a local PM and 2 developers and a software tester).

RESULTS & FEEDBACK

Can you share any outcomes from the project that demonstrate progress or success?

The Mindcraft team was able to come with alternative, creative but effective approaches to bring the project to a successful ending. Many future capabilities and possible approaches to reach even higher success rates have already been discussed for a V2 of the delivered product.

How effective was the workflow between your team and theirs?

Communication with Mindcraft is very easy. Jira, Confluence, Email, Slack, Zoom and other tools were actively used - Answers to questions or concerns came back very fast. The full team is capable of communicating in English on a very high level - both in speaking and understanding.

What did you find most impressive about this company?

We discussed, they offered and over-delivered. Impressive.

Are there any areas for improvement?

Nazar might have to take a BBQ training before our next project? (internal joke)

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

Research & Dev for Technology Services Company

“They were good—they delivered at the highest level.”

Quality: 
5.0
Schedule: 
5.0
Cost: 
5.0
Willing to refer: 
5.0
The Project
 
$10,000 to $49,999
 
Jan. - June 2019
Project summary: 

MindCraft provided research and development services, building a machine learning platform focused on Google Street View. The work was for a third-party client, testing an idea for an interactive roadmap.

The Reviewer
 
11-50 Employees
 
Vilnius, Latvia
Maksim Shilov
CEO, Vulmax
 
Verified
The Review
Feedback summary: 

The deliverables from the MindCraft team met the requirements perfectly, demonstrating the feasibility of the machine learning platform and demonstrating their ability to produce high-quality work. They were professional, technically sound, and skilled.

The client submitted this review online.

BACKGROUND

Please describe your company and your position there.

I’m CEO of Vulmax, a technology services company. We offer application development, software testing, back-office support, and other IT-related services, as we've also been developing a referral networking program, targeting companies and individuals who serve the same business areas as we do.

OPPORTUNITY / CHALLENGE

For what projects/services did your company hire MindCraft?

Our client’s business faced a challenge and they reached out to MindCraft to help solve it. They needed them to research and then build a solution to detect street signs on roadmaps.

What were your goals for this project?

To create a solution that can determine the geolocation of an object and its properties using Google Street View images, proving that the idea was valid.

SOLUTION

How did you select this vendor?

They were recommended to us by a good partner we’ve been working with for a long time. Our technical specialist contacted their manager and after a conversation, we decided to work together due to their professional and competent approach.

Describe the project in detail.

The idea was to build a product capable of passing through a specific area on the roadmap. Using Google Street View images, they created an interactive roadmap with designated streets/places where parking is allowed or not, as well as different time-bound and other data. Our role is as a referral partner in the business relationship.

What was the team composition?

We worked with a team lead, a data scientist, and a machine learning engineer.

RESULTS & FEEDBACK

Can you share any outcomes from the project that demonstrate progress or success?

The project is under an NDA, but I can say that this proof of concept, and the customer’s idea, was developed well.

How effective was the workflow between your team and theirs?

It was good—they delivered at the highest level.

What did you find most impressive about this company?

The team showed good approaches in business analytics, project management, and full circle machine learning development.

Are there any areas for improvement?

No, they did a great job.

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

Intent Detection for AI Engineering Firm

"We managed to build effective communication with the MindCraft team."

Quality: 
5.0
Schedule: 
5.0
Cost: 
5.0
Willing to refer: 
5.0
The Project
 
Less than $10,000
 
Mar. - May 2019
Project summary: 

Mindcraft ventured to create a proof of concept (PoC) to provide metrics for an AI program's intent detection. 

The Reviewer
 
11-50 Employees
 
Wilmington, Delaware
Sergey Perets
CEO, Artua
 
Verified
The Review
Feedback summary: 

Mindcraft demonstrated that pre-trained embeddings provide an excellent opportunity to detect user intent via short messages. They exhibited a substantive understanding of machine learning, and there was effective communication throughout. They delivered the desired results in a timely manner.

The client submitted this review online.

BACKGROUND

Please describe your company and your position there.

I’m the CEO of Artua design and development agency. We provide our customers with UI/UX development also mobile app development services.

OPPORTUNITY / CHALLENGE

For what projects/services did your company hire MindCraft?

To build an engine for user intent detection in short text messages.

What were your goals for this project?

To build PoC that will give us an opportunity to obtain possible quality metrics for intent detection.

SOLUTION

How did you select this vendor?

Thanks to the recommendations of the people who worked with MindCraft, also due to the strong focus in Data Science Consulting.

Describe the project in detail.

For the purpose of PoC we choose NUS SMS corpus that is available from Kaggle. We use English-language part. Pretrained ELMO model was used to obtain word and sentence embeddings.

Several intents were labeled in original dataset using active learning approach to minimize labeling efforts. We show that the approach provided reasonable accuracy of intent detection and may be scaled for emerging intents while switching to live message flow.

What was the team composition?

Two data scientists and data engineer.

RESULTS & FEEDBACK

Can you share any outcomes from the project that demonstrate progress or success?

We have proven that pretrained embeddings provide an excellent opportunity to detect user intents is short text messages.

How effective was the workflow between your team and theirs?

We managed to build effective communication with MindCraft team. If was a need, we can have met at any time.

What did you find most impressive about this company?

MindCraft showed their strong knowledge in Machine Learning Area and great focus on clients problem using BA approaches.

Are there any areas for improvement?

The organization structure of the team should be more understandable for the customer and more independent inside the MindCraft. We were forced to confirm each step with MindCraft’s CEO to begin the project phase. Anyway, we obtained the results needed in a short term.

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