XEURUM

Background

Our client needed a specialized tool, which will allow him to quickly deploy 50—100 websites a day. We had to create complex system, which will give an ability to setup website to specific server with random or exact theme, plug-ins, configure Cloudflare and NGINX settings.

Solution (what was done)

We have created a system which:

• deploy WP and Joomla sites to multiple hosters;
• add special set of plug-ins;
• install themes;
• automatically setup NS and IP configuration on domain registrars, hosters and Cloudflare;
• change parametres of Apache and NGINX;
• can quickly reinstall website or delete it.

We were making software requirements, design mockups, MVP and several enhancements after that. Current system is a web application. Back-end was made on PHP, Front-end was made using React and Yii2 Framework.

Background

Every day linkbuilders create lot’s of profiles on forums and websites. They fill profile details, upload avatars and warm up accounts, so admins won’t delet them and posts they create.

But for several tasks, there was a need to create lot’s of accounts on different forums and sites. And all those account should loke like if they were created by the same person. This person should have the same face, but pictures should look like if they were taken in different places and at different conditions. And those photos must be unique.

Solution and results

After small research, we have stopped on NvidiaStyle Gan 2 architecture. We have created prototipe, made several geenration and understood that this architecture can solve our problem.
We have created a client server app, so many users in company could use it. In user interface we have created control panel, which allowed linkbuilder to create male/female character and bunch of different photos of him.

Background

Our client has a very big marketplace with millions of pages with different categories of goods. He had a SEO department and copywriters, who were making SEO texts for those pages. But such manual approach requires a lot of money and is very time consuming. And department had it’s own limitation. People couldn’t write as many unique articles as platform needed so client was searching for ways, how to optimize it.

Tasks and challenges

When GPT-2 was released for public in 2019, we have started to make experiments with it. We have generated first portion of SEO texts and have published them. We were suprized with results. Text were indexed by google and we’ve got positive dynamic in search results positions.Despite of those results, GPT-2 out of the box had a lot of limitations that didn’t allow us to use it in way we want. It had lack of context memory. It was generating 2-3 sentences about computers and on 4th or 5th sentence it started to describe reign of the English queen. It was known limitation of GPT-2. Also, we had to finetune model for our own tasks. But anyway, we made a Proof-of-concept.

Solution (what was done)

We started R&D process. We were playing with other systems like ERNIE, BERT, RoBERTa and different models of GPT-2 (124m, 345m, 774m, 1558m). After series of experiments and finetunig process, we have found a way, how to generate long texts (3 000 symbols / A4 letter long) without loss of context. We have used GAN approach to control quality of generated texts and to filter unwanted content.

What the client got (results)

After 7 month of R&D and 3 month of development of MVP, client got system, which could generate thousands of SEO texts for marketplaces in very short period of time.
Each text was unique, we were not using any king of spining technologies like Article Forge use. SEO department got a unique and fully customizable tool for it’s needs.
One of the models was made available as a public Demo and anyone could try it.

Background
While developing Websites Deployer, we found that we need to monitor health of that enourmous amount of websites and servers. For this purpose it was desided to create monitoring system.

Solution and results

We developed a custom website health monitoring system that allowed the admin to monitore websites and set time periods for health checks. The system could automatically retrieve the list of websites from WebSites Deployer or allow the admin to add websites manually. In case of incidents, the system sent email notifications to the admin. Also it can send messages to selected telegram addresses. The admin could also view the history of each website.

Our solution helped the client to detect and resolve website incidents in a timely manner, ensuring the continuous availability of their websites.

Background

Our client owned network of websites and they were creating content for their partners. They needed a CRM system that could streamline the content creation process, enable effective writers management, and ensure timely delivery of high-quality content.

Solution and results

We developed a custom content management system that allowed the client to efficiently manage their writing team and allocate tasks based on capacity, ratings, and prices. The system enabled the client to handle 1200-1400 texts per month, exceeding their initial target of 500-800 texts per month. The client reported significant improvements in the quality and timeliness of the content produced. The project was completed within 6 months.