Technical debt often The unknown villain of the enterprise, the disruptive companies seeking to modernize as they realize it how A lot of “legacies” live in their stack. As with most types of debt, there is usually interest to be paid as well.
This is something that a startup in the UK started Application agent The company seeks a solution with a platform that helps organizations automatically re-engineer their legacy applications, preparing them for deployment in a new cloud-native home.
AppFactor was officially founded in mid-2021, with CEO and founder Keith Nelson It has only been fully worked on since January recently Close He says the pre-seed round was worth £1 million ($1.3 million).
While presenting on stage today as part of TechCrunch Disrupt’s Startup Battlefield, Nelson demonstrated AppFactor’s technology and laid out his startup’s mission in a space that’s ready for change. TechCrunch reached out to Neilson previously to get the lowdown on the scale of the problem as he sees it, and what — exactly — AppFactor is doing to address it.
Up Factor Team. Image credits: Application agent
Vision
To outsiders, some technical debt may be evident through exposure to bugs or slow systems. Or perhaps the amount of time a company takes to improve existing products and introduce new features.
Meanwhile, in-house workers may have a better idea of their technical debt when they see that their IT budget is disproportionately spent on maintenance versus building shiny new things. Data from consulting firm McKinsey suggests that technical debt could be responsible up to 40% Of the total corporate IT budget, while a Separate report on Stripe It indicates that developers spend an average of a third of their work week addressing issues with existing technology rather than writing new code.
But it is not always easy to get a clear picture of the condition level of a company’s technical debt, since it may span multiple domains and areas within the organization. This blind spot might include things like code that is too complex, duplicated, or downright bad; Lack of automated testing; security vulnerabilities; And overall bad design.
“The big challenge for organizations is that they have to build and engineer enterprise-class applications simultaneously [specific] “In time, business requirements and processes change the environments around these applications, and the applications and their dependencies evolve over time,” Nelson said.
Thus, technical debt is perhaps best viewed, as McKinsey points out, as a kind of “tax” that a company pays on all internal development focused on fixing a myriad of legacy technology infrastructure. This includes new libraries and frameworks, or integration points and dependency changes as companies adjust their stack. Ultimately, it comes down to a whole patchwork of complexity that builds up over time to create an unwieldy mess.
A typical example of a legacy enterprise application might include a legacy Microsoft SQL database; Some middleware layers; and the .NET front-end, which requires a mix of physical and virtual infrastructure to function. The running processes, libraries, dependencies, and overhead components that permeate the application and infrastructure will require significant manual work just to figure out what’s what, as they try to shift lift and shift to a more cloud-native model.
This, essentially, is what AppFactor intends to offer. It scans a company’s IT environment to identify all of its applications and their dependencies, physically “detachs” virtualized and hosted applications from their existing environment, and rebuilds each component and application layer into separate containers ready for their new home – whether that’s a modern cloud architecture like Kubernetes, or Managed database service.
“All of this is created and led by the product [AppFactor]“So you can quickly move your existing applications to the latest cloud technologies in a matter of days, not months and years,” Nelson said.
![](https://techcrunch.com/wp-content/uploads/2023/09/633c13150d380b39ac9a37de_01_dashboard-1-p-1080.png)
AppFactor: App Update Module Overview (scheduled for release in November). Image credits: Application agent
under the cover
AppFactor consists of three core components, including a scanner/analyzer that is deployed on servers to collect the data needed to detect their applications and their dependencies; the “coordinator”, which primarily controls the behavior of the scanner/analyzer, including the IP range and target systems; And the comprehensive AppFactor SaaS platform handles all your data analysis, machine learning (ML) processes, services that create visual mappings, containerization tasks, and more.
The company says it is working with some commercial clients, including a UK-based enterprise software company Civica. Until now, only the “discovery and evaluation” aspect of its platform has been commercially available. However, the company is also gearing up to launch its “App Update” module in November. This means that customers will have the ability to not only find the right candidates for modernization, and provide all the relevant reporting and analysis, but ultimately enact the transformation itself.
Perhaps one of the most interesting features of the platform — from a bells and whistles perspective, at least — is a tool that enables users to visualize application dependencies through a 3D visualization engine. Eventually, this could be used to visualize entire environments.
“Right now, it’s more about infrastructure and operations, but there’s clearly room to go deeper, which is what we plan to build,” Nelson said.
Oddly enough, AppFactor also makes this available for VR headsets, with the company demoing this functionality via Oculus at its TC Disrupt booth.
“One of the most difficult activities that can help eliminate risks up front [app] “Change is the ability to calculate, view and understand dependencies — whether that’s across infrastructure, architecture or code,” Nelson said. “This view is about being able to view the structure and anatomy of our application properties and interact with them in a granular and powerful way. Some of these systems are incredibly complex, with connections, libraries, files, services, processes and more happening in large places and across multiple environments, so this is a really powerful way to be able to understand Knowledge, validation and reaffirmation intuitively, which enables any future development of the application and its features.
![AppFactor: Visualize application dependencies and entire environments through a 3D visualization engine](https://techcrunch.com/wp-content/uploads/2023/09/Gif1.gif)
AppFactor: Visualize application dependencies and entire environments through a 3D visualization engine. Image credits: Application agent
Playing status
Existing application update tools are largely manual and therefore resource-intensive. This may involve using a command line tool such as Docker, which requires significant ongoing testing, and even then may not cover the full range of dependencies due to the manual nature of running the tool. The likes of Google Migrate for Anthos, which resulted from its acquisition of Velostrata five years ago, and AWS’s App2Container It makes it a bit easier for organizations to convert virtual machines (VMs) into containers. However, these programs are still very manual and command-line based, do not necessarily provide comprehensive visibility into dependencies and do not support applications based on physical infrastructure.
There are other similar services that also focus on helping companies move from monolithic software to microservices, Such as Vfunction supported by the project.
The ultimate goal of each of these services is to help companies reduce their technical debt and “keep up with the times,” albeit by adopting slightly different approaches along the way.
“We believe there are four pillars of technical debt — infrastructure, architecture, code, and dependencies,” Nelson said. “We also believe that there are many applications that are not suitable for microservices, so our vision is to let the attributes of an enterprise application dictate the optimal architecture pattern.”
Artificial intelligence agent
To achieve this, AppFactor says it is developing machine learning taxonomies to help create the patterns needed to transform more complex “multi-host” applications. Basically, it’s about creating “fingerprint” technologies to identify complex or custom applications they’re made of.
“We use a trained data model to build this, and it uses a number of attributes and data points that can help identify patterns in the application,” Nelson said.
Additionally, Nelson said they are experimenting with a number of other AI use cases, including large language models (LLMs) to create Hopes (A human-readable data serialization language for creating configuration files) for Kubernetes deployments
“We have some [other] “Future use cases are around code generation, but we’re not there yet,” Nelson added.