back to Services

Machine learning
and AI

Even though AI was founded more than 60 years ago, it has exploded very recently. This mostly has to do with the emergence of powerful hardware, cloud and mobile technologies that allow handling and processing of Big Data.

Lemberg team leverages these opportunities to develop Machine Learning (ML) algorithms that turn Big Data into meaningful information for the end users. Reach out to us to learn how your product can generate exceptional value powered by AI.

Our expertise

Lemberg's ML & AI Team combines academic Data Science knowledge and excellent IT engineering skills. 

Document icon

Project types

Activity trackers

Chatbots

Autonomous vehicles 

Smart wearables

Robotics

settings-work-tool

Technical expertise

Raw data collection and pre-processing

Mathematical modeling and neural network training

ML implementation (API, SDK, etc.)

Data science team

Experience in consumer, industrial and medical projects

Took part in academic and commercial research

Own tool-set for data analysis and cleanup

 

Data science team

Machine Learning team
Technologies we use
Roles & Responsibilities
Business Analyst

Defining business needs and goals

Requirements gathering and info structurizing

Data visualization

Data Scientist

Researching

Modelling

Finding patterns & relationships

Data Engineer

Collecting, managing, analyzing data

Building & maintaining infrastructure

High performance computing
 

Software Engineer

Software, Firmware and Hardware development

Testing

*The number of team-members varies depending on the project type and complexity.

How we work

Frame the Problem
Identify key business challenges and desired goals
Collect Data
Collect all data provided and find out additional data points
Evaluate Data Quality
Evaluate the quality of collected data to figure out how it will impact the training process
Preprocess Data
Structure collected data and perform data clean-up to improve accuracy
Use Chosen Machine Learning Algorithm
Select optimal machine learning algorithms, in case of neural network – design the topology, then train the network and perform results analysis
Implement Informational Model
Wrap up the data model in a library and then integrate it with a business product (app, backend or firmware)
Evaluate Modelling Result
Evaluate final results, iterate and continuously improve your data model
What you get
Research report

Brief information on the experiments carried out during the research, their methodology, the result obtained and their interpretation

Prototype

Implementation of developed algorithms in form of Prototype

Informational model

A model obtained as result of data processing with machine learning algorithms in form of neural network, configured fuzzy logic controller, mathematical formula etc.

Algorithm correction

Tuning and optimizing of algorithm for better performance

Collected datasets

Data sets contain the value of informational parameters of the modeling object and are used to construct and validate an information model of object or process.

Field test report

Report with results of comprehensive Field testing of the prototype and their interpretation

Algorithm description

Description of the principle of the functioning of solution

Description for patent registration

For the know-how found during the research, technical documentation for the registration of a patent for a utility model may be prepared.

Projects

Hand Hygiene Monitor

Hand Hygiene Monitor

Motion Gesture Detection

Gesture Detection

Horse Analytics

Horse Analytics

Contact us

Please tell us more about yourself and your project.

Lou Dutko
Chief Technology Officer
I consent Lemberg Solution Ltd collecting and using provided personal information as set out in the Privacy Policy.
x