NLP development for pharmaceutical company
Challenge: The client, a multinational pharmaceutical company wanted to refactor an existing system and improve its performance and analytical capabilities through AI.

Team: Python and Angular development team

Solution: Our team used semantic analysis with NLTK preprocessing to create different feature extraction and drug product labeling over already accumulated and classified еnquiry data. Preprocessed data were used to train Keras with TF engine, and evaluated versus built-in SVM algorithms in SKLearn. Results were accuracy near 80% for classification. ML components were integrated into the auto-assign pipeline based on product and team recognized in free text еnquiry.

Status: Successful completion of refactoring tasks and introduction of NLP module.
“Trust their advice. They’re knowledgeable and can create a product better than you imagined.”

N. Hauser, Technology manager, Pharmatech
React development for a Construction industry solution
Challenge: The client was a UK start-up company having its internal Python team. Their challenge was to find a matching senior enough team to work on the front end of the software.

Team:  ReactJS Lead and ReactJS senior developer, PM and UX

Solution: The solution included multiple interactions between users, both internal for the construction company, and external partners or subcontractors. This required a high level of security and attention to security and authorisation of users.

Status: Our team completed the project successfully. Currently, the client has hired an internal team to continue the project.
“We’re now managing the frontend in-house, but it’s easy for us to go into the code and understand how things are moving because of their high-quality development work.”

Delyan Ruskov, Structor.io
AI screening tool for a recruitment software
Challenge: Adding an AI tool for screening candidates in an HR start-up.

Team: Python and data science team

Solution: The first step in the process was to set up a process for collecting data in different formats from multiple sources. Once the data flow and organization was set, we used semantic analysis with NLTK preprocessing to extract the necessary information for each candidate. The ML part was further developed with Keras, TensorFlow and SKLearn. The output presented structured database of candidates.

Status: The module was successfully integrated.
"It is exciting to see have a partner who is not only technically capable but has also business expertise. Blagovesta has been a COO of a prominent HR company and her support in view of product design were of great help to us."

Samuel Johnson, Founder, Recruit
MVP development with AI for a start-up in manufacturing
Challenges: Developing a software which helps managers and engineers to analyze the performance of their production facility taking into consideration the combined output and performance of machines and people.

Team: 4 Python Django developers, 1 PM part-time

Solution: The solution was successfully developed to MVP phase and continued to include computer vision combined with tasks, procedures and read-only registers from PLC controllers. We developed a functioning AI software, which collects information from the machines through PLCs, analyses their idle time and thus present recommendations for improvement.

Status: Successfully completed the MVP phase.
“During the development process they were proposing solutions, advising us on every step we were in doubt, and we really felt them part of our team.“

G. Moore, Co-founder, FactorIoT
Stock market data tool integrated with Bloomberg
Challenges The clients wanted to improve its data analysis tools and support its team with an automation in their daily work.

Team: Python and DevOps team

Solution: We have built a tool integrated with Bloomberg service terminal to extract historical and real-time data for the NYSE. Currently, our team is working in tight cooperation with the client to apply numerical methods to compose different types of feature vectors and work with Keras to predict MTD / YTD prices based on previous period evaluations.

Status: Completed with ongoing development
“They are good with big data, really good. Trust their advice. They’re knowledgeable and can create a product better than you imagined.”

A. Bahtev, CEO, MNDB
Web platform for travel reservations exchange
Challenge: Building a secondary marketplace for travel reservations, based on the model of eBay for flights and holidays.

Team: PHP, Python, Vanilla JS, DevOps

Solution: The development involved structure of multiple roles and permissions, payment approvals and exchange of constantly changing information. The task required a highly experienced team of developers which we ensured for the client. Apart from the front-end and back-end development, we supported them with UX advisory on every step.

Status: Successfully completed with ongoing development and support
Childish created a much better website than what we had previously. They made helpful suggestions and thought about every detail and how it fits with the bigger picture. Their team displayed excellent product-management.

G. Stavreva, CEO, SpareFare.net
Development for online betting platform
Challenge: A tipster betting platform working with multiple APIs

Team: Python, DevOps. PM

Solution:  The task was a complex platform interconnected with multiple other sources of information flows through multiple APIs. Because of its complexity, multiple potential vendors had already rejected the project, however, our team did a great job, gave confidence to the client that the process will work fluently and it really did. All was done on time and within the quoted budgets.

Status: Successfully completed with ongoing development and support
Childish was a capable partner, communicating with intent and ensuring all requirements were understood. Their involvement and ability to work smoothly were key traits of the team.  They stuck to our initial budget, and we made some changes for parts that weren’t in the original project free of charge.

I. Antonov, Sportalaxy.com
SaaS Computerized Maintenance Management System
Challenge: The task was to develop a solution for task management in production companies combined with extracting data from production machines through sensors and PLCs.

Team: 4 Python developers, 3 Angular developers, QA, DevOPs (AWS) and PM

Solution: The solution was successfully developed to MVP phase and continued to include computer vision combined with tasks, procedures and read-only registers from PLC controllers. Currently using Python OpenCV & YOLO object detector integration from IP camera stream to extract workers' movements around specific industrial areas. We collect data first to perform supervised learning in later stages.

Status: Core modules complete, we work on future development. Currently, the software is being integrated into 3 production plants in Bulgaria.
“We knew the team even before Childish and this is why we have decided to entrust our key production machines to them”.

Eng. I. Ivanov, Plant manager, Rakovski
Educational platform for children
Challenge: Building a platfrom with exciting UX for the children and complex back end with multi-level users. Children play Q&A games on books that they have read and compete with other children.

Team: Python, DevOps. PM, UX

Solution:  The platform was created withing the tight deadlines and runs smoothly. During peak times of competition, it had around 100,000 people answering and working on the site, and the site responded successfully.   

Status: Since its development in 2018, we have developed multiple upgrades and new modules, a mobile application and the clients are more than happy and thankful for having us as a partner.
"They made everything we wanted, working cleanly and efficiently with no mistakes. They are really great proffesionals, great people and partners for us."

Iskra Djanabetska, Co-founder, Knigovishte.bg
This website uses "cookies". By continuing to use it you state your consent to that. I agree