Stermedia

5-Year Global QA Partnership – Full-Cycle Software Support by Stermedia

For over 5 years, Stermedia has provided end-to-end Quality Assurance support to a global e-commerce company, contributing to the development of desktop, mobile, and web applications. Our QA engineers seamlessly integrated into the client’s international teams, delivering both manual and automated testing (Selenium, Cypress), significantly improving product quality and streamlining release processes.

Thanks to our flexible approach, technical expertise, and clear communication, the client was able to scale teams quickly, implement changes effectively, and avoid delays in feature delivery. The project also included internal QA tool development and mentoring of client QA teams.

Technologies: JavaScript, Cypress, Selenium, Jira, TestRail, GitLab
Services: QA, manual & automated testing, QA tool development, CI/CD process support

AI Platform Takeover & Optimization for a US Tech Company

A Virginia-based tech company developing an AI-powered platform for content strategy and SEO automation engaged Stermedia to take over and enhance the project during a major team transition.

The original platform was partially manual, lacked proper documentation, and suffered from siloed team structure. Stermedia deployed a lean, highly skilled team covering backend, frontend, data science, DevOps, and QA.

We streamlined workflows, introduced daily cross-team collaboration, and began automating core processes. Our approach led to a smooth project handover, significant cost reduction, improved product stability, and a clear roadmap toward full AI automation.

The client continues to work with us to evolve the platform into a scalable, self-service SaaS product powered by machine learning and natural language processing.

Stermedia supported a fast-scaling same-day delivery company by enhancing its mobile application for couriers and integrating backend systems for operational efficiency. The client required external development expertise to improve order acceptance workflows, implement real-time, location-based task recommendations, and optimize user experience under time pressure.

We provided a dedicated software engineer who collaborated closely with the client’s internal developer, working in an agile environment. The project involved mobile development (React Native), backend integration (Node.js), and UX improvements, while ensuring compliance with App Store and Google Play policies. The result was faster delivery times, increased courier productivity, and a successful cross-platform app launch.

AI-Based Tender Filtering System for Enhanced Sales Efficiency

Stermedia developed a custom AI-driven solution to streamline tender processing for B2B sales teams. The platform automatically scrapes tenders from multiple sources, filters them based on user-defined natural language criteria, and generates concise summaries to support faster, data-driven decision-making.

Key Features:

• Dynamic, descriptive filters instead of static keyword searches

• Automated tender summaries (title, deadline, value, description)

• Integration with Discord, Microsoft Teams, and email

• Built with Python, FastAPI, PostgreSQL, and Kubernetes

• Enhanced by Large Language Models (LLMs) for learning and precision

Results:
The client reduced manual tender review efforts by up to 90%, saving time, reducing costs, and enabling teams to focus on the most relevant opportunities.

Project Name: Goldsaver - A Flexible Gold Investment Platform

Client: Goldenmark (Leading Gold Trading Brand)

Industry: Financial Services, E-commerce, Retail

Project Overview: Stermedia partnered with Goldenmark, a prominent brand in the gold trading industry, to develop Goldsaver, an innovative online platform that enables retail customers to invest in gold in small, flexible amounts. The platform allows users to buy gold gram by gram, at real-time prices, and accumulate it until they reach the value of 31.1 grams (1 ounce), which they can then collect physically or continue to invest further.

Key Features and Solutions:

• Custom Software Development: We built a seamless system combining backend technologies (Python, Django) with modern front-end tools (React, TypeScript), ensuring scalability and performance.

• ERP & System Integration: Goldsaver was fully integrated with Goldenmark's existing ERP, accounting, and warehouse management systems, enabling smooth data synchronization and streamlined business operations.

• Advanced Security Features: The platform includes advanced security measures such as two-factor authentication (SMS, physical keys), identity verification, and fraud prevention mechanisms to protect users' financial data.

• Dynamic Exchange Rate Handling: Real-time updates of gold and currency exchange rates, ensuring that prices reflect global market fluctuations and public holidays.

Challenges Addressed:

• Market Demand for Flexible Gold Investment: Before Goldsaver, customers had limited options to invest in gold without committing to large amounts. This platform offers an affordable, accessible solution for small investors.

• System Integration Complexity: The integration with Goldenmark's existing ERP and accounting systems was a challenge due to the need for real-time synchronization and compliance with industry regulations.

• Transaction Security: Implementing robust security measures, such as fraud prevention and multi-factor authentication, was essential to ensure the safety of customer data and investments.

Business Impact:

• Increased Customer Reach: The platform attracted over 80,000 new users, allowing Goldenmark to reach a broader customer base and cater to small investors who were previously unable to participate in the gold market.

• Improved Efficiency: The integration of systems and automation of processes resulted in significant operational efficiencies, minimizing the workload of accounting and warehouse teams.

• Enhanced Trust: The advanced security features built into the platform helped build customer trust, leading to increased loyalty and long-term relationships with clients.

Technologies Used:

• Backend: Python, Django, REST API, Celery (for asynchronous task management)

• Frontend: React, TypeScript

• Hosting: Cloud infrastructure ensuring reliability and scalability

Conclusion: Goldsaver is a revolutionary solution in the retail gold investment space, offering a flexible and secure platform for individuals to invest in gold. Through collaboration with Stermedia, Goldenmark has not only expanded its offerings but also optimized its internal processes, demonstrating how custom software development can address market challenges and create new business opportunities.

Our team collaborated with a private investor and an experienced doctor to develop an AI healthcare platform aimed at transforming access to primary care in Saudi Arabia. The platform allows patients to input symptoms, receive AI-generated diagnostic reports, and share them with healthcare providers, ensuring informed decision-making and seamless communication.

The project was delivered as a Minimum Viable Product (MVP) within a fixed budget and timeline, despite challenges such as prioritizing essential features and adapting to evolving requirements. Through design workshops and mockups, we ensured the platform met both patient and doctor needs. Currently in testing, the platform is set to revolutionize primary care with innovative telemedicine solutions.

AI-Powered Healthcare Solutions: Transforming Mental Health Screenings

Our team developed an innovative AI-driven healthcare solution designed to improve depression screening processes. This cutting-edge application leverages advanced technology to automate assessments, enhance diagnostic precision, and support medical professionals in delivering timely care.

Project Highlights:

• Comprehensive Modules:
  • Questionnaire Module: AI evaluates patient responses and body language to identify emotions.
  • Honesty Analysis Module: Algorithms detect sincerity through speech and facial expression analysis.
  • Depression Assessment Module: Combines insights to diagnose depression and assess severity.
• Agile Methodology: Flexible, iterative development ensured adaptability to evolving healthcare needs.
• Collaboration with Experts: Continuous input from doctors and specialists refined AI accuracy and relevance.

Results:

• Accelerated patient assessments.
• Enhanced primary care doctors' ability to detect mental health conditions.
• Reduced pressure on psychiatric specialists by improving initial screenings.

This solution embodies our commitment to leveraging AI for impactful healthcare advancements. Explore the case study to see how we’re shaping the future of mental health diagnostics.

About the Client

Our client is a rapidly expanding company in the automotive industry, specializing in used car rentals and leasing. Their mission is to provide competitive offers while ensuring operational efficiency and accuracy.

Project Description

The client faced a manual, time-intensive process for preparing vehicle offers, taking up an entire day to create only 40 offers. This process involved matching incomplete vehicle descriptions with industry catalogs, often leading to errors and inconsistencies. They approached Stermedia to develop an AI-driven solution to streamline their operations and enhance accuracy.

Objective

To automate and accelerate the process of creating vehicle offers, reduce errors, and provide a scalable, efficient solution that meets the client’s evolving needs.

Services Provided

• AI Development: Leveraging Generative AI to automate data retrieval and matching.
• Custom Software Development: Creating an intuitive dashboard for offer management.
• Business Analysis: Identifying processes for automation and optimization.

• Founded in 1953, Ferguson is a leading distributor of construction supplies in the U.S., operating with about 36,000 associates across 1,700 locations.

• Their challenge was data management. Employees dealt with vast amounts of information, and the Excel-based system was no longer sufficient.

• Their main goal was to replace the cumbersome Excel system with a cloud-based app that would simplify data management by consolidating information into a single, reliable source. 

• The initial business analysis for Ferguson's project established the scope and evolving requirements. Starting with basic sketches, we refined the project through multiple iterations. 

• Ferguson received a customized solution based on extensive UX research. Currently, we are preparing for deployment and initial employee testing at the client's site.

• Read the full case study here: https://stermedia.ai/how-to-streamline-construction-management-with-technology/

• Automatspec is a European leader in the B2B vending machine sector. The company has 38 branches, employs over 300 staff members, and offers over 5,000 active devices.

• The management recognized areas within the company that could be streamlined with AI, and our role was to identify them.

• Through a 20-hour workshop, we sought answers to the question: Can the quality of employee work and customer satisfaction be improved through AI?

• AutomatSpec received a list of specific AI technologies they could apply to their current processes to free up employee time and automate daily repetitive and manual tasks, including:

• An AI chatbot for the text communication channel
• A voicebot for handling telephone calls
• A summary generating application

• Next step? Implementation!

• Check the full case study here: https://stermedia.ai/how-to-streamline-corporate-work-with-ai/

• Dr Vaartjes is a Dutch dentist with a strong business mindset. He envisioned an app that would generate detailed dental treatment plans and provide real-time updates for patients. 
   
• Similar systems exist in the Netherlands but are outdated in visual and functional standards. Our task was to create an advanced yet simple tool for presenting a step-by-step treatment plan in a clear PDF format.
   
• The project involved coordination among the client, a design firm, and a project management company. An exciting challenge was dealing with complex data and the legal requirements of dental treatments. 
   
• We developed an interactive treatment planner with a visual editor. The app can automatically generate patient letters covering a detailed treatment plan.
   
• The first clients in dental clinics are now using the initial version of the application.
   
• Check the full story here: https://stermedia.ai/a-dental-treatment-planning-app/

• When new vehicles of an automotive giant came off the assembly line, they underwent tests that revealed technical problems, such as issues with electric mirrors or engines.
   
• The manufacturer's internal diagnostic system recorded these problems but only provided basic information about their occurrence without offering guidance on fixing them.
   
• As a result, employees had to rely on their knowledge and intuition, leading to a time-consuming trial-and-error process.
   
• To address this, we developed an app that logged faults and provided workers with concrete, data-driven assistance to resolve technical issues quickly and efficiently. 
   
• It automatically suggests likely solutions to employees when they encounter problems.
  
   
• Automated issue-solving system
• User-friendly dashboard
• Prioritized solutions
• Error analysis and reports
• Swift data handling
• User intervention option

• Anna, a team leader, approached us for assistance when she decided to transition from her current managerial role to a more strategic position.

• Her team was divided into several subteams. While there was occasional collaboration among them, they were a collection of individuals rather than a unified entity.

• She aimed to prepare them for enhanced self-management but couldn’t pinpoint the specific issues that might hinder the team's progress with her limited engagement.

• To find those issues, our Scrum Master, Hanna, delved into the team's dynamics and identified the core challenges:

• 3 months of observing and mentoring the data team

• Defining roles and assigning responsibilities to team members

• More self-management and open feedback

• Data team ready for the leader's role change
  
  Read the full story here: https://stermedia.ai/implementing-self-management-in-decathlon/

A European company wants to enter the local electric mobility market with a modern electric car rental platform.

The question is: is there a demand for it in this country? We conducted a comprehensive market study and competitive analysis to find out.
 

1.  Market Research
    

   We conducted a thousand interviews with potential users.

   The key takeaway was that there indeed is a market demand for this kind of system, and the local electric mobility market is highly receptive.
    

2. Competition Analysis
    

   We focused on the top 20 players in the electric mobility sector from four key markets.

   We focused on understanding their business models, data structures, and system functionalities to grasp the industry's current state.
    

3. MVP Architecture
    

   We developed the IT architecture of the MVP, focusing on insights from our research. 

   We paid special attention to the security due to the size of our client's company and the project's significance.
    

4. Prototype and Testing
    

   We created a service prototype and presented it to potential clients. 

   This phase marked the first real-life application of our theoretical research. The key takeaway: the experience should be unique and luxurious.
    

5. Business Analysis
   
   We developed key business cases involving detailed costs, revenues, and profits analyses.
   
   We helped the client understand how to organize logistics and business processes to launch sales and test the MVP concept.
    

6. Time for the Real Test!
    

   In less than three months, we created the first version of the portal, ready for market launch. 

   The MVP aims to verify the market in terms of actual interest in subscription and system usability for real users. 

   What happens next? We're eagerly waiting to find out! A 6-month pilot will soon begin.

Our client, a Dutch-Polish software company with over 18 years of IT development experience, supported one of the world's largest companies in the Testing, Inspection, and Certification (TIC) sector. They aimed to develop an app to replace traditional on-site evaluations by consultants.

• The concept was that customers would use everyday objects like a credit card or an A4 sheet as reference points to upload photos of the damage. The system would then automatically calculate the area of damage on the surface.

• However, the software company lacked the specific mathematical expertise to implement this idea. This is where we stepped in.

• Our task was to design a system that was accurate, efficient, and user-friendly, placing the responsibility for damage size estimation on the user. We focused on avoiding overengineering and utilized simple mathematical models.

• Our team consisted of two members: a doctor in mathematics specializing in algorithm development and a university professor who also served as a Python developer. Instead of creating a complex AI engine, they used straightforward mathematical models and classical problem-solving methods.

• Within just two weeks, they delivered a fully functional system. The application consistently provides precise analysis results in about one second, significantly reducing manual workload and improving customer response times. While the system is lightweight and straightforward, it is robustly prepared for future development.

Check out the full case study here: https://stermedia.ai/how-to-use-simple-math-models-for-major-business-challenges/

WMC IT Solutions, a Swiss company founded in 1981, specializes in IT services for the insurance brokerage sector and small to medium-sized enterprises. Their flagship product, BrokerStar, is a critical platform over 100 insurance brokers use to manage clients' insurance portfolios.

WMC needed to migrate BrokerStar from Symfony 1.8 to version 2.4 to maintain service excellence. The migration posed several challenges.

• The system had no initial documentation and thousands of unique in-app use cases.
• The outdated Symfony version was nearing the end of its support period, posing security risks and compliance issues.
• Additionally, BrokerStar needed modernization to meet growing market demands and ensure scalability.

We partnered with WMC to handle the entire migration process.

• We managed intricate systems of commissions, referrals, client retention, billing, invoicing, and API data access.
• We rewrote the application using the latest Symfony framework, retaining all original features and adding new modules.
• We introduced a new translation system for four languages and reorganized data management for improved usability.
• We also enhanced the permissions system, making it more granular and faster.

The migration was a success. 

• BrokerStar now runs on Symfony 2.4 with better performance, data security, and compliance. 
• The application is multilingual and accessible in four languages, including English.
•  Data organization and user permissions are more efficient and user-friendly.
• We also maintained and expanded technical documentation and unit/functional tests.

Post-migration, we continue to serve as WMC's technological partner, providing maintenance, error fixes, updates, and new feature implementations. We ensure BrokerStar remains compatible with various external systems and maintains high performance and security standards.

Full case study:

https://stermedia.ai/updating-a-complex-broker-system/

The client was a company specializing in location-based services and smart workspace solutions.

In 2017, they approached us to redesign their app, aiming to enhance its appeal and functionality. Their goal was to update the existing internal designs to attract new potential clients.

Our initial task of app redesign quickly expanded. We transitioned from a design-focused role to becoming an essential part of their technological strategy, effectively serving as an external technology department.

Over the course of seven years, we assisted the company in developing a more user-friendly mobile app, driven by user feedback. This new solution replaced their old physical beacon system.

Currently, our focus is on maintenance and ensuring everything runs smoothly, allowing the client to concentrate on their product vision and strategic direction.

Full case study: 

https://stermedia.ai/from-redesign-to-tech-partnership/

Public Health Department

One of the European Public Health Institutions oversees sanitary safety in the country, handling various activities from hygiene and environmental health to epidemiology. 

This institution also regulates the quality of food and consumer products, ensuring their safe introduction to the market.

Manual Notification Handling

When a food manufacturer introduces a new product to the market or changes the composition of an existing one, they must send a special notification to the institute.

Officials receive and manually verify the correctness of these notifications by checking:

• Whether the product’s chemical composition meets specific legal requirements.
• Whether the manufacturer has informed about the product’s origin.
• Whether the label is translated into Polish.

And many other criteria required by national and European law.

After checking all variables, officials either approve the correct application or reject it if it fails to meet any requirement.

Our system automatically detects and signals any irregularities and generates a set of recommendations for officials: it advises whether a submission can be accepted or not, highlighting the reasons for rejection.

Automating and Simplifying the Process

The Public Health Institution faced the challenge of automating and simplifying the notification verification process for new products.

The existing process was time-consuming – the staff, despite their high competence, could only process several applications per day, while tens of thousands were received annually.

Additionally, the flood of notifications was irregular; half of the year’s notifications could arrive in one month, creating more delays.

The consequences were significant. The prolonged queues risked allowing products with unapproved substances to enter the market, endangering consumer health and jeopardizing the reputation of producers.

There was a need to replace manual labor with an automated system to significantly increase efficiency and allow employees to utilize their potential better.

A Machine Learning solution was needed.

As artificial intelligence specialists, we were approached by Sotfiq – our shareholder and primary technology provider for the Public Health Department – to design a system to address this issue and elevate the institution to a new technological level.

Understanding the Problem Thoroughly

We began the project with consultations with the Institute’s employees.

Initially, we met to establish a common communication language. This was important because they used many terms that meant something different in ordinary language and within the department.

And vice versa – since officials don’t work with new technologies daily, we also had to explain and standardize our terminology.

The second part of the workshops focused on profoundly understanding the institutes’ work environment. We went step by step through the notification handling process, identifying areas for improvement along the way.

We also delved into relevant legal regulations to navigate the process effectively, from European journals to national legislation.

Only after establishing common, uniform terminology and a complete understanding of the process and legal environment did we move on to proposing solutions.

We conducted a week-long brainstorming session with employees, where we suggested appropriate technologies for various process stages.

The employees then evaluated their practicality and compatibility with the institution’s operational realities, and a final vision of the solution was crafted.

Designing the System

We designed a solution that automates the label verification process using OCR (Optical Character Recognition), CV (Computer Vision), and NLP (Natural Language Processing).

OCR and CV technologies automatically recognize label contents, including text, graphics, and ingredient tables.

NLP is used for further analysis of label contents; after being processed by OCR and CV, it enables a deeper understanding of the content, including identifying threats and irregularities.

We developed a series of automatic verifiers that classify information on labels based on established criteria.

Results

The initial results came during the workshop phase: our external perspective on the processing showed how much the process could be improved, not just technologically.

The workshops provoked employee discussions about the process, from common naming to universal practices in customer service. Together with our client’s employees, we standardized and organized the knowledge within the organization.

More importantly, our client was provided with a comprehensive project specification encompassing a complete technical solution, the necessary infrastructure for its maintenance, and the essential team for its implementation.

The specification could serve as part of the tender documentation or as the foundation for development with their current technology suppliers.

However, our workshops and solution proposal pleased the employees so much that they decided to commission us to create this system.

Wakeb had a business idea to build an innovative image recognition solution for the future. We verified their vision and created a ready-to-go solution that they can now present to private and public investors.

Wakeb Technology 

Wakeb, headquartered in Riyadh, is a pioneering force in the region’s artificial intelligence and machine learning fields.

Their future-oriented projects are backed by deep research and international support, ensuring their delivery sets industry benchmarks for innovation and quality within the established timelines.

Our shared appetite for tomorrow’s tech and easy-going business savvy got our partnership rolling. 

Crafting Products for Tomorrow

Wakeb envisioned a futuristic Image Recognition Platform that could autonomously evolve and shift through various data types, including spotting animals, vehicles, and people.

The founders sensed the business potential for such tech, seeing its application across diverse sectors—from industrial production to advanced security systems, right through to Smart City solutions.

With Wakeb’s team fully committed to their ongoing projects, they needed a tech partner to turn the business vision into a tangible product that could take on the world commercially.

They required an MVP that was more than just a proof of concept for hypothesis testing; it needed to be a durable foundation for further development.

We aimed to design something robust and scalable from the outset, not just a temporary fix.

Quite a challenge, no?
 

Refining the Scope: Business Vision vs. Technical Viability

From a futuristic idea to a finished product is quite the journey—first, we had to figure out if the concept was feasible and, if so, with what resources.

Crucially, every cutting-edge AI project needs a unique R&D plan because there are no ready-made solutions. We’re making discoveries as we go. 

That’s why we brought in Jan. He’s been working with AI since 2016 and is excellent at putting all the pieces together to see the whole picture. He took a deep dive into the project’s assumptions and started the research, exploring everything from existing commercial solutions to academic papers on image detection.

After 20 hours of R&D, Jan returned with a verdict: without a clearly defined end-user, crafting Wakeb’s perfect vision wasn’t possible at this stage.

Instead, he suggested simplifying the concept to a model with a basic API that could easily integrate and detect three key object classes. This kind of product would give us a strong foundation to build on and improve in the future.

This product would give us a solid foundation to build on and improve. With this MVP, Wakeb could confidently enter the market to seek investors or end customers.

The Wakeb crew liked that sound, so we rolled up our sleeves and got down to development.
 

Four Weeks of Crafting The Future

For four weeks, we were crafting the solution; we introduced our Scrum process, complete with planning sessions and daily retrospectives. Wakeb was open to following our framework, and we started working together on the solution.

Wakeb provided the business know-how, reviewed each product version, and corrected the project trajectory.

We handled the coding and ensured everyone from the business end to the tech team stayed in the loop. 

Week 1: Discovery

• Looking into the latest tracking techniques.
• Gathering data sets.
• Testing out chosen models.

Week 2 and 3: Implementation

• Putting the features into a script.

Week 4: Presentation

• Creating an API that bundles the components. 
• Setting up a process for easy deployment on various devices.
• Making unique visualizations for easier object identification.
   

From Concept to Practical Solution

Wakeb’s received a one-size-fits-all solution ready to roll out in all sorts of places, from factory floors to airports and even smart city tech that monitors the crowd and traffic.

This model’s flexible; with a bit of work, it can fit right into different market spots. It’s good to go and can be customized wherever it fits.

Just needs the right moment to shine. 🙂

Looking ahead, we also proposed Wakeb a bunch of ideas to make the app even better, like making it work on specific devices or as a web app.

We’ve sketched out a game plan for growing the project further with heatmaps, image classes, and an intelligent feature that learns from your pictures or descriptions.

So, what’s in store for Wakeb?

With their brand-new AI Image Recognition tool ready to be showcased to investors and governments, we’re much closer to finding out!

Our portfolio you can see at https://stermedia.ai/case-studies/

We have experience in AI/Machine Learning and Deep Learning services for

• Industry/Automotive
• HR
• Medicine
• and other business like Startups