Techstack Ltd

PRODUCT: A set of desktop utilities and graphic editors under a single brand. 

CHALLENGE: A large project required maintenance of high development quality and testing speed.

SOLUTION: We started automation testing from scratch in a massive and complex graphic design system. Our main tasks were building a testing framework as quickly as possible and covering most of the code with tests. One more problem was the flood of bugs reported and revealed after bug fixes. The processes needed to be more flexible to withstand the growing bug reports. 

• The team has grown from 0 to 8 members and included manual and automation QAs.
• We introduced double-stage Kanban sprints, where the QA team tests developed features while the development team worked on new features in parallel.
• We optimized the smoke test to save time before releases.
• We introduced testing in different environments (Trunk, Staging) to ensure impeccable release quality.
• We divided the test team into sub-teams each focused on their product to allow engineers to concentrate better and ensure better quality.

OUTCOME: Over the course of our cooperation, the product has become financially self-sufficient. Parallel development and testing sprints ensured an uninterrupted software delivery.

• Unit tests count: 5,000+
• Automated tests count: 2,500+
• Releases we took part in: 50+

ABOUT THE PRODUCT: An open sales engagement platform that integrates content, communications, and actionable insights to improve business results. 

CHALLENGE: The product needed a module for gathering, analyzing, and presenting data about the efficiency of sales campaigns.

SOLUTION: Warming up clients’ interest in the product, adding more value and extending subscription packs are essential for retention of existing customers.

We built a handful of high-efficiency integrations of product analytics tools:

• Tableau allows for extended personification of analytical data—know your customer by their company, title, location and track their flow.
• Gainsight PX provides inputs directly from end-users. The interviews you can acquire through this tool help outline efficient and customized feature adoption strategies.
• FullStory enables the recording of anonymized end-user flow to better understand their needs and identify the root causes of their behaviors.

OUTCOME: We slowed down client attrition and helped grow the platform’s client base. All the improvements led the product to acquisition at the beginning of 2021 by one of the leading players in sales enablement.

ABOUT THE PRODUCT:

A set of desktop utilities and graphic editors under a single brand.

CHALLENGE:

The maintenance of an enterprise platform that incorporates 5 global brands with dozens of integrated products. We needed to beat the low performance, develop custom features, and solve instantaneous technical issues.

SOLUTION:  

We designed a serverless backend to handle high loads and the majority of business flows, such as licensing, authorization, data management, and AI. We also covered the system with unit, integration, and auto-tests, which unlocked continuous delivery.

Custom features we built and continue to maintain include a cross-platform advertising system, administration panels for brands’ products, payment processing integration for advertising systems, marketing reporting, and single sign-on. The custom code we integrate into the platform follows similar patterns to speed up development and testing time.

If necessary, we migrate from one database to the other, update ad localization keys, do whatever may prevent downtime, or improve user experience and system quality.

OUTCOME:

• We have implemented a CQRS pattern
• We have made the infrastructure fully auto-scalable
• We have introduced development by microservices

We have covered the system with unit, integration, and auto-tests, which unlocked continuous delivery.

ABOUT THE PRODUCT:
Face- and image-matching service for quick access to photos taken at mass events. 

CHALLENGE:
People generate thousands of photos during mass sports events. Our team was to build a service for automated face matching and photo search, storing, and sale.

SOLUTION:
We took a few steps to reach our goal. Here are the most important of them:

Stage 1: POC

We built a proof-of-concept to verify whether matching faces in selfies was feasible. The algorithm we built used AI and deep neural networks.

Stage 2: MVP

We developed a web application allowing users to search photos by face, store, get access to photos and sell them.

Stage 3: PSP integration

We added watermarks and payment integration to enable photo purchasing.

Stage 4: Going serverless

When the platform scaled and attracted more users, server costs increased. We reduced them by implementing serverless architecture. 

Stage 5: Maintenance

We continue to improve and maintain the product and work closely with our partners on the product roadmap.

IMPACT:
For each stakeholder group, we provide value:

• Organizers don’t need to store and provide access to photos themselves;
• Participants now can quickly find their pictures;
• Photographers get access to their target audience and can sell photos to them.

TECH STACK: Azure, .Net, Angular, Javascript, CSS, HTML

ABOUT THE PRODUCT: A platform for quick access to services in hospitality that instantly delivers a client’s demand to staff by real-time request routing.

CHALLENGE: Identify the problems that users face in the hospitality sector and search for cases where process automation can bring value to the business.

SOLUTION: The Techstack team built the product from scratch, starting from the idea. We have built the following apps and units:

• Two cross-platform mobile apps built with PhoneGap and migrated to React Native
• A progressive web version of the app on reusable components between React Native and React
• A web dashboard build with Angular 7
• A real-time backend built with Node.js and Socket.IO.

OUTCOME: The product we designed and built implements the idea of being where the client is.

Enabling an automated interaction with clients wherever they are, we improved customer service and added value to the hospitality industry.

PRODUCT: Portal for сollecting and storing patient history before visiting a doctor, using a Survey module. It ameliorates the processes of organizing the work of the hospitals.

CLIENT: A leading provider of breast information management solutions, including options for planning, sharing results, patient navigation, and more.

CHALLENGE: When we started our cooperation with the partner, the Survey module allowed doctors to get answers to all predefined questions before the patient’s visit, but it had limited capacity. It was necessary to give a unique assembly for each individual hospital. Also, the technologies needed to be updated to improve productivity and availability.

SOLUTION: We started by building a new web portal for patients’ access, which was integrated with central data storage (Visual FoxPro). Its functionality has multi-step patient verification.

We created the stack of the Portal and Survey module, using ASP.NET MVC Framework, and updated the system to .NET Framework 3.5 and wrote a new frontend with AngularJS. This means that we updated the whole stack.

We rewrote the code of the Survey module and implemented a new approach. The functionality of the module consists of a Survey consumer, Survey provider and Survey library.

VALUE DELIVERED:

• Serviceable Portal
  We recreated the Portal and Survey module using the actual technologies. This influenced the system positively, made it more convenient and faster.
• Configurable surveys for patients
  The questions are completely customizable to any facility, so it is possible to gather the data that is vital to different medical practices. The data is encrypted, so the patient’s information stays safe and secure.
• Accessible additional forms
  In addition to the history questionnaire, centers can also migrate other necessary forms to an online portal. This can include privacy and consent forms, disclosures and any other forms that require a patient signature.

TECH STACK: .NET, C#, ASP.NET Web API, SignalR, AngularJS, PWA.