The Augmented Architect – unleashing the power of ChatGPT

Consider ChatGPT as a virtual assistant, dedicated to streamlining your workflow and taking your architecture to the next level.

An Augmented Architect?

Gone are the days when IT architects had to rely solely on their own expertise and intuition to make critical design decisions. Today, we live in an era where technology is constantly evolving. It is crucial for IT professionals to stay ahead of the curve, through technology and sector monitoring, networking, and engaging in the Community of Practices and ChatGPT represents another valuable opportunity for IT architects to become “Augmented Architects“.

But what exactly is ChatGPT, and how can it help IT architects do their jobs better? Put simply, ChatGPT is a state-of-the-art AI language model that can generate human-like text based on a given prompt. For IT architects, this means that they can leverage ChatGPT’s advanced capabilities to streamline their work, improve their accuracy, and ultimately enhance the outcomes of their projects.

How does ChatGPT enhance the work of IT architects? Let’s examine three examples to understand their impact.

1. Architecture design assistance

IT architects constantly face complex design problems that require swift and efficient solutions.

ChatGPT can assist in the design process by providing suggestions for architecture patterns, best practices, and trade-off analysis:

• Pattern suggestions: ChatGPT can provide suggestions for commonly used architecture patterns that are relevant to the specific problem the architect is trying to solve. This can save architects time and effort in researching and identifying suitable patterns.

• Best practice recommendations: ChatGPT can recommend best practices based on previous experience and industry standards. This can help ensure that the design is scalable, secure, and efficient.

• Trade-off analysis: ChatGPT can provide a trade-off analysis to help architects weigh the pros and cons of different design options. This can help architects make informed decisions and choose the best course of action.

Here is a concrete example of a trade-off analysis, conducted with ChatGPT:

“Imagine an IT architect is tasked with designing a new e-commerce platform for a company. What are the possible architecture scenarios, considering factors such as cost, complexity, scalability, and performance?”

After several deep dives and interactions with ChatGPT, here was its answer:

“When it comes to designing a new e-commerce platform, IT architects have to weigh several factors such as cost, complexity, scalability, and performance. Two common architecture scenarios to consider are composable commerce architecture based on microservices and monolithic architecture. Let’s explore the trade-off analysis between these two options:

Option 1: Composable Commerce Architecture based on Microservices

In this approach, the e-commerce website is built using microservices, each responsible for a specific function such as product catalog, shopping cart, payment processing, etc. This approach offers greater flexibility and scalability, as new microservices can be added or removed as needed. However, it also involves a higher level of complexity, as each microservice must be managed and maintained separately.

Option 2: Monolithic Architecture

In this approach, the e-commerce website is built as a single, integrated application. This approach is simpler to manage and maintain but may be less flexible and scalable than the composable commerce architecture.

IT architects can benefit from ChatGPT’s fast and effective support in the design process, freeing up time and resources to delve into scenarios and concentrate on strategic matters.

2. Component selection

The selection of the right components for an IT architecture can often be a time-consuming and challenging task. With ChatGPT’s ability to analyze large quantities of data and provide informed recommendations, IT architects can more easily select the most suitable components for their projects, saving time and effort in the process.

I asked ChatGPT:

“Let’s consider a scenario where an IT architect is tasked with searching for the right front-end framework for a company website. Can you recommend a set of solutions to consider, with their related advantages and disadvantages?”

ChatGPT responded with:

1. React has a relatively low learning curve, but it may require more resources compared to the other two options. Additionally, it benefits from a thriving community that offers extensive support and resources.
2. Vue.js is known for its simplicity and flexibility, making it a popular choice for building user interfaces. However, it may come with a slightly steeper learning curve compared to React.
3. Angular offers a comprehensive solution for building web applications and has many features and tools out of the box. It also has a steeper learning curve compared to React and Vue.js, but once you understand the framework, it can be a powerful tool for building complex web applications.”

Such answers from ChatGPT can serve as a starting point for architects who are unfamiliar with the topic, or as a way to accelerate the work of those who are already knowledgeable in the field.

3. Technical documentation

Technical documentation is a critical but often time-consuming aspect of an IT architect’s work.

ChatGPT can offer valuable support by streamlining the documentation process through its ability to process large amounts of information and understand technical language.

With ChatGPT’s help, architects can more easily summarize and structure technical data, allowing them to allocate more time to higher-value activities.

As an illustration: you will find a  summary, generated by ChatGPT, at the very end of this article.

Here’s another concrete example of how ChatGPT can assist in technical documentation:

Let’s say we need to document the architecture of a new cloud-based application. The architect can provide ChatGPT with the necessary technical specifications, design documents, and any other relevant information. ChatGPT can then assist in generating a high-level summary of the architecture, outlining the main components, relationships, and dependencies.

This summary can serve as a useful starting point for the architect, who can then use it to generate diagrams (ChatGPT cannot) or create a more detailed technical document. In this way, ChatGPT can help IT architects save time and effort in writing technical documentation while ensuring that the information is accurate and well-structured.

This allows architects to focus on more value-added tasks, such as identifying and resolving technical issues or exploring new technologies to drive innovation.

Limitations

While the benefits of ChatGPT for IT architects are clear, it is important to recognize that the technology is not flawless. ChatGPT’s performance is limited by the data it has been trained on. In my testing, I also encountered some errors in the context that required restarting the interaction with ChatGPT.

ChatGPT now officially warns you that your messages may be read by others (including the previously mentioned issues with context), so we need to be careful not to disclose any confidential information.

Additionally, as with any AI technology, it is crucial to critically assess the validity and relevance of the information produced by ChatGPT. There may be instances where its suggestions are partially, or even entirely, inaccurate. The way you formulate your questions can mitigate this risk. In a nutshell, ChatGPT is not a substitute for experts, but it can support them if used effectively!

What about sustainability?

The use of large language models such as ChatGPT poses a challenge in terms of sustainability, because the training and deploying of these models require a significant amount of computational resources and energy, resulting in a large carbon footprint.

Not only does the training process consume vast amounts of energy, but the deployment and ongoing use of these models in production also consume significant resources. Given the environmental impact of using large language models, it is imperative to explore and adopt more sustainable approaches for AI development and deployment.

Implemented practices, such as using renewable energy sources and optimizing the use of resources during training and deployment are necessary, and there is still much to be done in this area. It is crucial for the AI industry to focus on sustainability and reducing the carbon footprint of AI systems.

My conclusions

ChatGPT can become a valuable asset for IT architects in their daily work, offering numerous benefits.

Whether it be assisting with architecture design, component selection, or technical documentation generation, ChatGPT has the potential to transform the way IT architects work.

It can ultimately increase their productivity, efficiency, and effectiveness.

I’m convinced that technologies such as ChatGPT have the potential to serve as virtual assistants for not only architects but also many other professionals in various industries.

So why not try ChatGPT, and experience the benefits for yourself?

Summary (generated by ChatGPT)

This article, co-authored by Pascal Espinouse and ChatGPT, explores the potential of using advanced AI language technology, ChatGPT, to boost the productivity and success of IT architects. ChatGPT can act as a virtual assistant and provide support in areas such as architecture design, component selection, and technical documentation.

Throughout the article, real-life examples are presented to illustrate the impact ChatGPT can have in these areas. While the benefits are clear, it is important to acknowledge the limitations of using ChatGPT and emphasize the need for critical evaluation and decision-making by the IT architects.

Ultimately, the article concludes by stating that ChatGPT is a valuable tool for IT architects to become “Augmented Architects”, but also for many other professionals looking to work more efficiently and effectively.