Developing with artificial intelligence

Artificial intelligence has become an integral part of software development. We use AI where it creates real added value and are also aware of the challenges involved. Instead of just relying on hypes such as generative AI, we integrate customised AI solutions that have been successfully implemented in practice for our customers – always with the aim of solving complex problems and achieving sustainable success.

More than just GenAI

For us, artificial intelligence is a practical tool that helps companies overcome real challenges. Our team of experts develops customised AI strategies that perfectly fit the needs of your project. We not only offer AI consulting, but also support you in finding the right time and the best way to use AI so that companies can get the most out of it.

1. AI-powered document processing for the insurance industry

soxes has developed a mobile application for BBT that uses AI-based functions to improve the quality of document scans, especially on older devices. The AI ensures smooth processing of insurance documents by minimising sources of error and reducing manual effort. Thanks to the isolated instances, the app can also be used as a white-labelled product for other insurers, which reduces development costs. Further information can be found here: BBT Software for Insurance.

2. Innovative HR technology: AI for analysing job applications

Assessment Gate AG has developed a modern HR solution for the automated application process in collaboration with soxes. The existing «Prisma World Solution» has been technically and functionally expanded to support companies in recruiting personnel. With the help of an AI model to determine suitability, candidates can be assessed efficiently and based on data. A REST interface enables improved user-friendliness, while the system also offers individualised development reports for employee dialogues. You can find more information here: Assessment Gate AG reference.

3. optimisation of production quality with artificial intelligence

For Emmi Group, soxes developed an AI-supported prototype for quality assurance in yoghurt production. This uses Microsoft Custom Vision to check whether the correct lids are being used on the yoghurt pots. The «fruit images» are analysed automatically and errors are corrected immediately. Thanks to the agile collaboration, soxes was able to implement an effective system within just six weeks that optimises production and strengthens Emmi’s quality standards.

4. Smart fault detection in household appliances

Problem: Electronic appliances such as washing machines, televisions or hoovers can exhibit various faults. Correctly interpreting these faults and associated messages and providing quick solutions can be a challenge for service departments.

Our solution: We have developed an AI-supported fault detection database that analyses huge volumes of fault messages and derives statistically relevant patterns from them. The AI can not only recognise the most common errors, but also provide a precise analysis of how often these errors occur, why they occur and which solutions have proven successful in the past. In doing so, AI takes into account different levels of analysis and model development to deliver precise results.

This type of wealth of experience that AI builds up helps companies to deal with troubleshooting more efficiently and offer their customers a significantly improved service. As a result, problems can be solved faster and in a more targeted manner, reducing both downtime and costs for the customer.

5. AI-supported image recognition for nature lovers

Problem: People who are out and about in nature want to find out quickly and easily which bird species they are observing. Conventional identification books are impractical and not always up to date.

Our solution: With our AI-based image recognition software, users can take a picture of a bird using their smartphone camera, which is sent to a cloud-based artificial intelligence in real time. This evaluates the image and recognises the bird species using training data from thousands of bird images.

This technology is based on advanced inference, which makes it possible to recognise patterns and draw conclusions from new data. Inference refers to the ability to combine existing knowledge with current information to gain meaningful insights. Our image recognition software can be used flexibly and individually in numerous industries and areas.

6. free text queries for medical practices

Problem: A large amount of information is managed in medical practices – from patient files and prescriptions to appointments and schedules. Practice staff often need to retrieve this data quickly and efficiently.

Our solution: Using free-text query software, doctors and their teams can ask questions to the database in ‘natural language’. The AI converts the queries into precise database queries and quickly delivers the required information. Whether it’s calling up prescriptions, viewing reports or managing appointments, the software makes the entire process more intuitive and efficient.

This solution can also be extended to other industries where complex data needs to be managed and made quickly accessible. AI for business has many applications here, from analysing medical data in healthcare to optimising processes in industry and improving customer service in banking.

7. Automated actions from emails

Problem: Manually processing emails can be time-consuming – especially for recurring tasks such as payments.

Our solution: AI software analyses incoming emails, recognises payment requests and automatically triggers the necessary actions. Today, technologies such as machine learning and natural language processing play a crucial role in helping companies achieve their goals. This saves time and reduces human error.

Additional use cases:

• Appointment scheduling: Automatic creation of calendar entries from appointment requests.
• Document forwarding: Attachments are forwarded to the relevant departments.
• Task creation: Emails are converted into tasks in project management tools.
• Reminders: Automatic sending of reminder emails for important deadlines.
• Customer support: Enquiries are forwarded directly to the support team.
• Reporting: Status reports are automatically extracted and integrated into the reporting dashboard.