Modernize Your Business With AI

AI, ML, NLP ALGORITHM IMPROVEMENT

AI SOFTWARE ENGINEERING AND DEVELOPMENT

OUR STORY

VESOLV is a software AI/ML solutions development lab, which was established in 2015 by two friends, MIPT alumni. The founders have 15 years of experience in IT software development for real business.

Now our team includes 20 people, and we are passionate about creating complex software solutions that provide a unique competitive advantage to our customers.

…AI Lab

From startups to large enterprises….

….AI, big data analytics and computer vision.

Core industries

AL/ML/NLP
modules for IT solutions

Scoring system

Project analysis

Recommendation system

Risk management

Voice of Customer

collect information about all communications between a company and its clients from various communication channels (text + voice). Perform classification, extract entities, build analytics.

Predictive analytics

apply mathematical analysis and statistics to predict structure and volume of demand

Voice of Employee

collect information from employees of a company about improvements of a processes

Performance & optimization

code and execution analysis including analysis of hardware limitations, code refactoring

Streaming and real-time data processing

build Real-Time OLAP with “delta” of several seconds

Text analytics

NLP, text processing, entity extraction, classification, aggregation

CASES

EXTRACTION OF MACROECONOMIC INDICATORS FROM TEXTS AND THEIR SEMANTIC LINKS FOR SENTIMENT ANALYSIS

Problem: There are progress reports and planned activities as well as projections and plans for the future. Unstructured data should lead to a more structured view - texts in tables-facts.
Solution: First, markers (indicators, quantitative and qualitative characteristics, institutions, activities, social object, etc.) and then relations between them (forecasts, implemented changes, current situation, etc.) are highlighted. Types of relations in this case get tonal color - negative, neutral, positive.

HOW TO AUTOMATE THE CALCULATION OF THE INSURANCE BONUS? SOLUTION IS CUSTOMER SCORING?

Problem: Potential the insurance premium amount consists of 3 main parts: theft risk premium, total risk premium and damage risk premium. The amount of the damage risk premium is calculated based on the average premium for the brand/model (determined on the basis of the average expenses of the insurance company on indemnification of losses for the respective vehicle) and the "accident risk factor". which is determined individually for each insurance policy taking into account the peculiarities of the vehicle and its drivers.
Solution: Calculation of individual level of accident risk (accident risk factor) on the basis of vehicle data, owners' history, accident history of the vehicle, fines, owners' data, etc.

CUSTOMER PROBLEM ANALYSIS: DEFAULTS AND LOAN DEFAULTS

Problem: bank customers often do not repay consumer loans taken from the bank in the first two years when interest payments are highest. The bank has to reduce the risks of non-payment at this time as much as possible, i.e. cut off at the stage of making a decision on granting loans to insolvent clients.
Solution: Forecast of client default in the first and second year on the history of client transactions

HOW TO AUTOMATE VEHICLE INSPECTION WHEN BUYING INSURANCE?

Problem: Potential fraudsters can deceive an insurance company when they do a photo inspection of their car on their own, for example, by replacing photos with older ones that do not yet have recent damage to the car. Thus, when buying insurance, they can report the damage they have actually sustained before buying the policy and repair it for free.
Solution: detection of fraudulent photos (old, from the screen, layouts, in the photo editor, similar cars) based on image/video stream analysis.

TECHNOLOGY STACK

AI/ML

* Algorithms improvement, data science and data mining, mathematical analysis and simulation (statistics, mathematical analysis, clustering, regression, anomaly detection, machine learning)
* Neural networks, deep learning (TensorFlow, Theano, scikit-learn, Keras, PyTorch, Caffe, MATLAB, MatrixNet, H20)
* Semantic networks, graphs and ontologies (Neo4j, GraphDB, Apache Jena, Stanford Protégé, OWL, RDF, Networkx)
* Selected implementations of gradient boosting (CatBoost, XGBoost, lightGBM)
* Text libraries - NLP (NLTK, Gensim, spaCy, GATE)
* Video and computer vision libraries (OpenCV, PCL, YOLOv3, object detection, face recognition, albumentations)
* Geographical location analysis (Osmapi, GeoPy)
* Custom data processing libraries and general mathematical apparatus (pandas, NumPy, SciPy, Mathematica, MATLAB, statsmodel, fbprophet, arch)

Mobile development

* Android
LANGUAGES: Kotlin, Java, JavaScript
FRAMEWORKS: Koin, Kodein, RxJava2, Kotlin Coroutines, ConstraintLayout, OkHttp, Ktor, Retrofit, React Native, Exoplayer, Navigation, Lifecycle, Firebase
DATA STORAGE: SQLite, Realm
* IOS
LANGUAGES: Swift, Objective-C, JavaScript
FRAMEWORKS: RxSwift, SwiftLint, OpenGL, Firebase, CoreML, Core Data, WatchKit
DATA STORAGE: Realm

Frontend

LANGUAGES: JavaScript
FRAMEWORKS: Vue.js, React, Angular, VueX, Redux, Bootstrap, RxJS, Apollo, Electron, Cypress, Socket.IO, Web Assembly
MARKUP: HTML5
STYLES: CSS

DevOpps

TECHNOLOGY: Docker, Kubernetes, Linux Containers, AWS, Jenkins, Grafana, Prometheus, Loki, Terraform Jira, Jenkins, Github, MS Team Foundation, Agile/Scrum

Backend

LANGUAGES: Go, C/C++, Python, Java
FRAMEWORKS: Spring, Echo, Flask, Spark, Gval, Gorilla, Asyncio, Aiohttp, Socket.IO, Apache Jena, Apache Kafka, Apache Camel, ActiveMQ, ServiceMix, Apache Solr, Apache Lucene, Sphinx, Google OAuth, Yandex OAuth
DATABASES: Redis, MySQL, PostgreSQL, ClickHouse, Oracle, MS SQL, PostgreSQL, Cassandra, GraphDB, Neo4j
* Distributed computing: Apache Hadoop, Apache Zookeeper, Apache Ignite, UIMA DUCC
MESSAGE BUSES, QUEUES: Kafka, NATS

Implementing and improving Odoo ERP (Open ERP)

* Development of new and modification of existing Odoo modules
- more than 3 years experience
- work with all versions, starting from 10
- experience of core modules deployment and modification
* Odoo technical support and maintenance
- deployment of various configurations
- industry-specific and general purpose modules
- third-party applications

HOW WE WORK

STARTUPS

Bring your application to market. 4 steps from idea to working MVP

  1. From vision to clear technical requirements. Through a set of product workshops we help you shape the product, identify ecosystem, set priorities and design a road map.
  2. UX/UI design. User flows are the heart of your product and the main key to customer satisfaction. At this stage we help visualize product concept and ensure right usability.
  3. Development. It is the riskiest stage for every project, and not all manage to pass this stage with reasonable outlays and time. To avoid development pitfalls our team works closely with the customer using agile methodology backed up by a detailed milestone structure.
  4. Iterate. Once your product goes to the public and starts to gain traction, we help you to reach product-market fit. There is nothing more permanent than temporariness. We are here to support you with changes and the right tech people.

SMBS

Everything you need to increase efficiency or build new vertical integrations.

  1. Project design. Revision of existing solution and product workshop to identify clear technical requirements, right ecosystem and project milestones. We complete this step with a development road map, which includes roles and full communication flow.
  2. UX/UI design and detailed planning stage. Core aim of this stage is pretty wise and simple – agree on the detailed project plan and deadlines.
  3. Build prototype. This stage is the main one, which involves helping to build the final product right from the first time. Testing of concept may help in avoiding mistakes and development pitfalls. Prototype is the right visualization of user experience.
  4. Development and Q&A. This is a coding stage, which is based on the clear and detailed road map while our project managers facilitate core advantages of agile approach.
  5. Launch and maintenance. We uphold warranty obligations for the delivered product and provide technical support, which includes minor changes base on short-term and long-term subscription.

ENTERPRISES

Enhance your growth and efficiency with the cutting-edge AI powered web and mobile applications

  1. Just as no enterprise is like each other, no application is a twin.
  2. Our dedicated enterprise team works on various types of applications to streamline enterprise growth.
  3. While we keep core process from project design to launch and maintenance, we specifically consider important enterprise needs like:
  • Enterprise grade security
  • Scalability
  • Security and specific legal compliance
  • Existing IT eco-system
  • Application life-cycle and cost of support

Feasibility studies for large enterprises is an essential part of the LeanTech approach and help to ensure that required resources are in place, or ROI projected in line with expectations.

Our Partners