DEV Community: Hans Reinl

Building a Metric Program That Actually Works

Hans Reinl — Thu, 19 Dec 2024 11:17:19 +0000

At LeadingEng in Berlin in November this year, I had the opportunity to lead a Table Talk on a topic that has been a significant focus of my work lately: building and implementing effective metric programs. It's a subject that sounds deceptively simple, but as I've discovered, it's a journey filled with complexities and invaluable lessons. This experience has resonated with my reflections on transitioning from engineering to management, as detailed in my blog post From Engineering into Management, where defining clear objectives and measuring progress are paramount for success. Just as in managing teams, managing metrics requires a strategic and thoughtful approach.

Like many organizations, we initially treated "metrics" as a buzzword. We knew we wanted to be data-driven, but the practical application was hazy. Product teams had their metrics, but engineering felt like a different beast. Were we even measuring the right things?

One of our biggest hurdles was the sheer diversity of our engineering landscape. We had teams working on iOS and Android apps, others building web applications with a mix of programming languages and frameworks. This meant we couldn't just use a one-size-fits-all approach. We needed a flexible framework that could adapt to different development processes and technologies while still providing comparable data across teams.

"Start somewhere" is good advice, but starting without a clear direction can lead you down a rabbit hole of irrelevant metrics. We needed a framework to identify the key indicators that truly reflected our team's impact.

Enter Engineering KPIs: Cutting Through the Noise

This is where Engineering KPIs became our guiding light. By focusing on a small set of clear, measurable metrics aligned with our team's goals, we were able to filter out the noise and focus on what truly mattered.

One specific example stands out: our initial approach to measuring errors. Like everyone else, we tracked the number of errors. But simply counting errors isn't enough. Not all errors are created equal. Some have a negligible impact on users, while others cause major disruptions. So, we took a more nuanced approach. We defined custom metrics that captured the errors with the biggest impact on our users. For example, we started tracking "crash-free users" as a percentage, which gave us a much clearer picture of user experience than just raw error counts.

The most valuable Engineering KPIs are those that directly support key product metrics. For a streaming platform like ours, metrics like video loading speed and buffering time are critical. Improving these, directly contributes to viewer engagement and satisfaction.

Metric	Value (Seconds)	Change (%)
Average Loading Time (Before)	20	-
Average Loading Time (After)	9	-55%

Table 1: Average Video Loading Time Before and After Optimization

From Data to Action: Turning Insights into Impact

Collecting data without acting on it is a waste of time and resources. The real value comes from using those insights to drive change. We implemented a two-pronged approach:

North Star Metric: A long-term, aspirational goal that aligned with the overall business objectives.
Short-Term Goals: Realistic, achievable targets set within 3-month cycles. This allowed us to iterate quickly and adapt to changing circumstances.

Regular reviews are essential. We established monthly meetings with our teams to discuss progress, analyze data, and adjust our metrics as needed. This continuous feedback loop ensured our metrics remained relevant and impactful.

North Star Metric (Product)	Engineering Metric	Target (Next 3 Months)	Initiatives/Actions
Improve User Retention	Average Time to First Interaction (TTFI)	Reduce by 20%	- Optimize database queries. - Implement caching strategies. - Improve API response times.
	Error Rate (Specific User-Impacting Errors)	Reduce by 10%	- Implement more robust error handling. - Improve automated testing coverage. - Invest in better monitoring and alerting.
	Deployment Frequency	Increase by 50%	- Implement CI/CD pipelines. - Automate testing and deployment processes. - Improve code review practices to facilitate faster releases.

Table 2: An example for engineering metrics with a North Star Goal

Building Buy-In: Metrics as a Team Sport

Metrics aren't just an engineering concern. Getting buy-in from other departments is crucial. We found that adopting OKRs (Objectives and Key Results) helped align metrics across the organization. This fostered a shared understanding of goals and how each team contributed to the overall success. This collaborative approach resonated strongly with the importance of cross-functional collaboration I highlighted in my blog post on The crucial role of an engineering manager in a company's success.

Metrics can be a double-edged sword. They can reveal inefficiencies, but they also celebrate successes. By fostering a culture of data-informed decision-making, we framed failures as learning opportunities and celebrated achievements based on quantifiable impact.

The journey to a successful metric program is ongoing. But by focusing on the right KPIs, taking action on the insights they provide, and fostering collaboration, we can empower our teams to achieve real, measurable business impact.

How GenAI Transforms the Landscape of Front-End Development

Hans Reinl — Tue, 28 Nov 2023 09:03:07 +0000

The transformative potential of Generative AI (GenAI) is revolutionizing the tech industry, and its impact is particularly profound in the realm of front-end development. This groundbreaking technology holds the key to automating mundane tasks, enhancing productivity, and unlocking new frontiers in design and development, enabling developers to create exceptional web experiences with greater efficiency.

Let's explore how GenAI will change the landscape of front-end development as we know it today.

Harnessing the Power of Generative AI: A Practical Example

To illustrate the transformative power of GenAI, let's explore its application in a React development project using ChatGPT.

ChatGPT demonstrated its ability to generate UI components, providing multiple variations with unique styles and functionalities. This allowed for rapid exploration of design options and selecting those that best complemented the user experience.

ChatGPT also assisted in crafting user-friendly interfaces by analyzing user flow and suggesting intuitive layouts, interactive elements, and clear error messages. This streamlined the process of creating a visually appealing and easy-to-navigate interface.

ChatGPT's High Points:

Overarching Plan Generation: ChatGPT generated a comprehensive plan for the development process, outlining the necessary files, API calls, and code generation.

ChatGPT's Areas for Improvement:

Outdated Examples and Libraries: ChatGPT sometimes provided outdated examples and libraries, requiring compatibility updates and increasing development time.
Maintaining Libraries: Some suggested libraries were not actively maintained, raising potential security and stability concerns.
Integrating AI-Generated Elements: Integrating disparate AI-generated elements into a cohesive and functional application proved challenging, requiring careful curation and refinement.
Contrasting ChatGPT and Bard: Approaches to React Code Generation

A comparative analysis of ChatGPT and Google's Bard chatbot in generating code for this application reveals distinct approaches. Bard consolidates the request into a single JavaScript file, aiming for a streamlined approach that reduces file management complexity.

While Bard's consolidated approach offers potential benefits in terms of code organization, it currently necessitates more refinement in prompting to ensure seamless integration with existing codebases or by creating this new application. Employing more precise prompts can potentially enhance the results generated by Bard, allowing developers to guide the AI towards their specific requirements.

Regardless of the approach used, the responsibility rests with the developer to ensure seamless integration of all generated code components.

Addressing the Backend Integration Challenge

Neither ChatGPT nor Bard currently provides a comprehensive solution for integrating generated code with the backend. The generated code lacked the necessary components for handling backend interactions, API calls, and data management. This highlights the need for further development in these areas.

v0: A Promising Option for Component Generation

v0, a tool developed by Vercel, offers a promising approach to code generation that complements the capabilities of ChatGPT and Bard. v0 excels at generating frontend components, providing developers with a variety of options to choose from.

Additionally, v0's user-friendly interface, with features like code copying, community example sharing, and forking capabilities enhances the development workflow. While v0 currently focuses on frontend development, its potential to integrate with backend services and provide a more comprehensive code generation solution warrants further exploration.

Conclusion

The AI-powered code generation tools currently available offer a range of benefits for React development, particularly in generating frontend components and providing a starting point for code development. However, the challenge of integrating the generated code with the backend remains a critical area for further development.

As these tools continue to evolve, developers can expect to see advancements in both code generation capabilities and integration support, leading to a more streamlined and efficient development process.

The future of front-end development lies in a symbiotic relationship between human creativity and machine intelligence. Developers who embrace GenAI will find themselves equipped with a powerful toolkit that can streamline their workflow, enhance their productivity, and unleash their creativity. As GenAI continues to evolve, we can anticipate even more sophisticated tools that further empower front-end developers to craft the next generation of web experiences.

The Crucial Role of an Engineering Manager in a Company's Success

Hans Reinl — Mon, 20 Nov 2023 19:22:44 +0000

In the ever-evolving landscape of technology, companies rely heavily on their engineering teams to drive innovation, deliver high-quality products, and maintain a competitive edge. However, engineering excellence doesn't happen in a vacuum. It's the result of careful planning, effective leadership, and a strong engineering culture.

At the heart of this success lies the role of the Engineering Manager, an often overlooked yet indispensable role that plays a pivotal role in shaping a company's technological trajectory. These individuals are the unsung heroes of tech, the visionaries who guide their teams towards achieving ambitious goals and creating groundbreaking products.

Disclaimer: In this article I speak about the role of an Engineering Manager, while in some contexts it might be referred to has Team Lead or even Head of.

Engineering Leadership: Navigating the Tech Landscape

Engineering Managers are not just managers; they are leaders who inspire, motivate, and empower their teams to reach their full potential. Their contributions span a wide spectrum, from setting clear strategic objectives to fostering innovation and resolving conflicts.

The role of the Engineering Manager

Vision and Strategy: Engineering Managers serve as the compass, setting the course for the engineering team and aligning their efforts with the company's overarching goals. They craft a clear vision that motivates and inspires engineers, ensuring that everyone understands their role in achieving the company's objectives.

Team Development: Nurturing the team's talent is paramount to long-term success. Engineering Managers act as mentors and guides, helping engineers develop their skills, advance their careers, and reach their full potential. They invest in training, encourage knowledge sharing, and recognize individual contributions.

Resource Allocation: In the dynamic world of technology, making the most of available resources is crucial. Engineering Managers carefully balance competing priorities, allocating resources effectively to ensure innovation, maintain existing products, and address immediate issues. They make tough decisions, prioritizing projects, distributing workloads, and investing in both talent and technology.

Conflict Resolution: Engineering teams often face challenges and disagreements. Engineering Managers are the mediators, facilitating open communication, resolving conflicts constructively, and ensuring that the team functions harmoniously. They address issues promptly and transparently, maintaining a positive and productive work environment.

Accountability: Engineering Managers uphold accountability within the team, ensuring that everyone takes ownership of their commitments and delivers high-quality work within agreed-upon timelines. They establish clear expectations, provide regular feedback, and hold team members responsible for their contributions.

Engineering Culture: The Breeding Ground for Success

A company's engineering culture is the bedrock of its technological achievements. It's the shared set of beliefs, values, and practices that shape how engineering work is conducted. A strong engineering culture fosters innovation, collaboration, and continuous improvement.

Innovation and Experimentation: Engineering Managers cultivate an environment that encourages innovation and calculated risk-taking. They empower engineers to explore new ideas, challenge assumptions, and experiment with cutting-edge technologies.

Continuous Learning: The technology landscape is constantly evolving, and a culture of lifelong learning is essential to keep pace. Engineering Managers support ongoing education, encouraging their teams to stay up-to-date with the latest trends and technologies.

Open Communication: Effective communication is the lifeblood of engineering. Engineering Managers foster open and transparent communication, both within the engineering team and with other departments. They encourage regular feedback, promote idea sharing, and break down silos.

Diversity and Inclusion: A diverse engineering team brings a broader range of perspectives and fosters innovation. Engineering Managers champion diversity and inclusion, creating a welcoming and inclusive environment where everyone feels valued and has equal opportunities to contribute.

Quality and Delivery: A strong engineering culture prioritizes delivering high-quality products on time. Engineering Managers instill a sense of pride in craftsmanship and a commitment to meeting deadlines without compromising quality.

The Driving Force Behind Tech Success

Engineering Managers are the unsung heroes of the tech world, the driving force behind a company's technological achievements. They are not just managers; they are leaders, visionaries, and mentors who play a pivotal role in shaping a company's culture and steering it towards success.

As your company navigates the ever-changing tech landscape, remember the critical role of your Engineering Managers. They are the architects of your technological future, the ones who will guide your company towards innovation, growth, and success.

The AI Future of Frontend Development

Hans Reinl — Mon, 06 Mar 2023 16:45:56 +0000

The world of programming has been evolving at a rapid pace in recent years, and the integration of AI into the development process is one of the most exciting advancements. AI tools such as ChatGPT have made it easier than ever to outsource programming work, and this has significant implications for the future of frontend development. As a result, it's vital to explore how AI is impacting the programming field, specifically frontend engineering.

In a recent episode of my podcast, Working Draft (in German), we delved into the impact of AI on engineering and discussed the future of working in the engineering field with a particular focus on frontend development. During this discussion, I shared my vision of where we might be heading in the coming years. This led me to write this article, where I can expand on some of the ideas that have been floating around in my mind for quite some time.

open.spotify.com

As we continue to embrace AI tools in programming, there are many questions that we must consider. For example, how will this impact the way we work? Will it change the nature of frontend development? What are the implications of this for businesses, developers, and end-users? How does the job of a frontend developer transform through AI? These are some of the critical questions that we'll explore as we delve into the future of frontend development with AI.

The Sigmoid Curve of AI

In his video titled "I tried using AI. It scared me.", Tom Scott offers his perspective on the potential impact of AI in the coming years. He suggests that we may be at the beginning of a sigmoid curve for AI, which could disrupt industries similarly to how Napster disrupted the music industry in the early 2000s. It made the CD obsolete and made way for music streaming as we know it today. The disruption of AI has the potential to significantly alter the way we work, live, and interact with technology. Maybe it even will make parts of engineering obsolete.

As AI tools become more prevalent, we can expect to see a greater level of automation in engineering. This will include automating tasks such as page creation, component development, and much more. However, it's essential to note that the role of each individual working in the field of engineering will be to act as the gatekeeper of the output of these tools. Engineers will be responsible for utilizing and leveraging these tools to their fullest potential.

With this increased automation, the job of engineers will likely shift towards a more strategic focus, with a greater emphasis on problem-solving and decision-making. As a result, engineers will need to develop new skills to effectively collaborate with AI tools and incorporate them into their workflow to create value even faster. While this change may seem daunting, the potential benefits of working alongside AI tools are vast and could lead to significant advancements in the field of frontend development.

AI stands for Artificial Intelligence and refers to computer systems that can perform human-like tasks such as learning, problem-solving, pattern recognition, decision-making, and language understanding.

– Thanks, ChatGPT, for this definition of AI

The Three Stages of AI

Artificial intelligence as such as been used as a term for a multitude of tools and approaches to work.
Generally, AI is broadly divided into three stages:

Artificial Narrow Intelligence (ANI): The chess program that can win against all human intelligence, think Siri, Alexa - fully available today.
Artificial General Intelligence (AGI): The AI is capable of performing comparable to a human: conversing fluently, making a coffee, get employed at a job and doing it well. We are not there yet, scientists think we will need a couple of more decades to fully unlock this level, maybe 2070.
Artificial Super Intelligence (ASI): At this level, an AI will learn from itself and will become more intelligent than human life, think sci-fi - it's in the distant future.

In our Working Draft episode on ChatGPT and AI tooling, Peter, Vanessa and I formed a different set of layers AI tooling is capable of doing:

Optimizing Input (Recontextualize): This layer involves optimizing input based on the model and requested prompt for a given context. For example, if a user provides a long text, an AI tool could be used to summarize it and extract the key information. Another example is translating a text into another language or turning a list of achievements into a text. This layer is useful for automating repetitive and time-consuming tasks, allowing engineers to focus on more complex challenges.
Reciting Information (Researching): This layer involves reciting information that the model is trained with and putting it into a given context. An example is an excerpt of a historic event and comparing it to another event. In more engineering terms, this layer allows AI tooling to find similar cases from what it has learned and put them into context. For instance, an AI tool could help engineers write Unit Tests based on a given input by finding similar cases that have been tested before. Another example is getting an answer to what the RegEx for parsing an email address is, by searching for similar patterns that the AI model has learned.
Bringing Together Data (Synthesis): This layer involves bringing together multiple data points and forming something new. For example, generating a music song based on the style of a genre of images - we know how to analyze images for their genres and could potentially build a song, bringing it together today is at least difficult. This layer is particularly useful for generating creative outputs, where AI can help augment human creativity by suggesting new ideas and combinations of ideas. However, this layer is also the most challenging, since it requires a high level of understanding of different data types and how they can be combined effectively.

Now let's see how these three layers can be put into action with frontend development.

From Design to Implementation to the User

A foundation for almost any website or application today is a design system in any form, at least describing a theme and re-usable components. Let's focus on building a design system using AI first.

Starting with a fresh design system, insights and recommendations for optimizing it can be gained by examining existing design systems and their construction.

AI tooling can be utilized to generate a design system by leveraging the synthesis layer to bring together multiple data points and generate an enhanced version. A potential tool could be fed with data on various design elements, such as color schemes, typography, and component interactions, as well as examples of existing design systems. The AI tooling would then use this input to generate a new, cohesive design system that adheres to the specific requirements and constraints provided. We will talk about it below.

A tool can be used to recontextualize the input based on the requested prompt, such as generating variants of a button based on different contexts, and ensuring that the output is tailored to the specific context it's used in.

So AI is a shoo-in to help wield a design system, and to participate in the creation of a design system’s assets. But there are many many many rocks that haven’t been turned over yet.

– Brad Frost, Design Systems in the Time of AI

Building a Design System

Theming

Building a design system isn't complicated. Just ask ChatGPT. Florian tried to generate a Design System using ChatGPT. I tried this simple request to build a design system for a website for running shoes and provide me with the color palette and typography. The output is quite useful.

Components

Taking this as a foundation for a design system, adding some design to create the initial components, or even easier, asking a tool like Midjourney to generate these, isn't a big step.

While I have made it work to generate a decent output for a button component, this is just a matter of experience, crafting the right commands and the machine to get the right data input. Let's assume, in a first step, you train your model on a bunch of existing design systems you can find via Design Systems Repo or similar.

Then you craft the initial component, let's say a button, based on the style input generated by ChatGPT using the tool until you are at the level of what you want. Now you can use this button as an input for your next request, for example a card component.
Generating any consecutive component will always match your design style. The AI does the heavy lifting.

The tool Uizard automates the process of generating designs and components from mockups. In his 2021 article, Tony Beltramelli discusses the underlying AI technology that powers Uizard and enables it to transform images into editable designs.

Design to Code

To further streamline the frontend development process, engineers can use tools like Fronty to generate code from their designs. This AI tool can create websites from mockups and code, eliminating the need for engineers to write every line of code by hand.

Once the code is generated, engineers can focus on separating it into manageable components, ensuring that each piece is optimized for performance and meets the specific needs of the project. By using AI to automate the code generation process, engineers can save time and focus on the more complex aspects of development, ultimately resulting in faster and more efficient project delivery.

One example of this efficiency is the possibility to generate code for multiple platforms at once. A tool is capable of providing output as a Swift UI component, as well as for React or another library for the web.

Generating Web Pages

On a web page, you want to display some matching data. New page-types get automatically generated based on the data, we see this with fully auto-generated, personalized experiences done by recommendation engines today.
Bringing this to the next level, an AI can generate endless pages with dynamic page types based on the input data you provide, for example user signals (analytics) and your businesses KPIs.

The Role of an Engineer with AI

Here is the gist of it:

Collaboration: Work with AI
Judgement: Control AI output
Critical thinking: Focus on high-level tasks
Custom solutions: Integrate AI
Ethical responsibility: Ensure ethical AI use

Let me explain this an in more depth.

As AI continues to become more integrated into frontend development, the role of an engineer will shift towards a focus on collaboration and strategy. Engineers will need to be skilled at working with AI and crafting the right prompts to generate the desired output. However, it's essential to note that the output generated by machines will only be as good as the data it consumed. Therefore, engineers will also play a crucial role in controlling the output of these tools, ensuring that the generated output is tailored to the context it's used in and that it's helpful in achieving the desired outcome. Moreover, as an engineer, your role will move away from knowing specific frontend frameworks to knowing specific Machine Learning tools.

For example, consider the generated components for a design system. Engineers will need to evaluate the necessary variants of a button in a given context that the AI tool needs to generate. They will also need to assess whether the generated variants are useful in the context they are used in, and ensure that the actions triggered by the buttons are properly mapped to the interactions with them. In essence, engineers will act as the critical bridge between AI tools and the final product, leveraging the power of AI while also bringing their expertise and judgement to ensure the optimal outcome.

In addition to collaborating with AI tools and controlling the output generated by them, engineers of the future will likely spend more time focusing on higher-level tasks that require critical thinking and decision-making. With the more mundane and repetitive tasks being automated, engineers will have the opportunity to focus on more complex challenges and use their skills and expertise to solve more significant problems. This will lead to the fact that the role of User Experience Designers and Frontend Engineers will move closer together.

This could include tasks such as designing complex systems, evaluating and selecting the most appropriate tools and technologies to use, and optimizing performance and scalability. Engineers will also need to understand how to integrate AI into these systems, evaluating the potential impact and benefits of different AI tools and techniques, and developing custom solutions to meet specific needs.

Moreover, engineers will play a vital role in ensuring that AI tools are used ethically and responsibly. They will need to be aware of the potential biases that can be introduced by AI tools and work to mitigate them. They will also need to ensure that AI tools are used in ways that align with ethical standards and best practices.

Overall, engineers of the future will need to be highly skilled, versatile, and adaptable, able to work collaboratively with AI tools while also bringing their unique expertise and perspective to the table. By doing so, they will help to drive innovation, enhance productivity, and create more effective and efficient frontend development workflows.

Reasons to Strive for this Future

There are several compelling reasons why striving for a future where AI is more integrated into frontend development could be highly beneficial.

Firstly, it could significantly enhance productivity and efficiency. By automating mundane and repetitive tasks, engineers could focus on more complex challenges and use their skills and expertise to solve more significant problems. This could lead to faster development cycles, shorter time-to-market, and ultimately, a more competitive and innovative industry.

From theme generation to componentisation to code to multiple platforms

Let's talk about the multi-platform example from above again: Imagine you want to build a design system that can be used across multiple platforms, such as the web, iOS, and Android. Today, this would require writing code multiple times for each platform. However, with AI tooling, these adaptations can be done automatically. By building a design system once in a single language, engineers could export the code and let machines handle the task of transplanting it into several languages for different platforms. This would result in massive time and cost savings, allowing engineers to focus on more complex challenges and innovations.

Secondly, it could help to address the shortage of skilled talent in the frontend development industry. With AI tools doing some of the heavy lifting, engineers could be trained to work more efficiently and effectively, allowing them to take on more complex challenges and develop a broader range of skills.

Thirdly, it could help to reduce the potential for errors and inconsistencies in frontend development. By automating tasks such as generating code and components, AI tools could reduce the likelihood of human error and ensure that development work is consistent and high-quality.

Finally, integrating AI into frontend development could help to drive innovation and improve the user experience. By leveraging AI techniques such as natural language processing and computer vision, engineers could develop more intelligent, responsive, and personalized user interfaces that better meet the needs and preferences of users.

Overall, a future where AI is more integrated into frontend development could bring significant benefits, enhancing productivity, addressing talent shortages, improving quality and consistency, and driving innovation in the industry.

Beyond Frontend Development

In addition to using AI tools like Fronty for design to code automation, there are many other possibilities for using AI in frontend and backend engineering. For example, ChatGPT can be used to automate backend tasks, such as generating system diagrams, optimizing database designs, and creating service meshes. With the help of AI, engineers can quickly and easily generate complex infrastructure diagrams and database designs, saving time and effort in the development process. Additionally, AI can be used to automate tasks like testing and deployment, enabling engineers to focus on higher-level problem-solving and innovation. Just take GitHub Copilot as an example. It "uses the OpenAI Codex to suggest code and entire functions in real-time, right from your editor."

Ask the machine, it will give it to you.

So What does it Mean for the Near Future?

The future of frontend development is exciting and rapidly evolving, with AI technology poised to have a major impact on the industry. As Brad Frost has noted in his aforementioned article "Design Systems in the Time of AI", there are already examples of how AI is automating aspects of the job, such as removing backgrounds from images. As AI technology continues to advance, it's likely that more and more of the frontend engineering process will be automated. This will not only lead to greater efficiency, but also a shift in the skill sets required for frontend developers. As such, developers who stay on top of the latest technology and continue to develop their skills will be best positioned to succeed in the industry.

While it's clear that AI will play a major role in the future of frontend development, there is still much to discover and explore. We can expect to see a variety of new tools and technologies emerge in the coming years, many of which will be built on AI principles. For example, ChatGPT and Midjourney are just two examples of how AI can be used to streamline frontend development processes. As more developers explore the potential of AI and incorporate it into their workflows, we will undoubtedly see a range of new innovations that will continue to shape the industry.

However, it's important to note that the human element of frontend development will never be entirely replaced by machines. While AI can automate many tasks, it cannot replicate the creativity and ingenuity that comes from human designers and developers. As such, it's important for developers to embrace new technology while also staying true to their own unique creative vision.

In conclusion, the future of frontend development is sure to be exciting and full of opportunities for innovation. As is with so many other roles. Let's see how rapid we are pacing towards this exiting future.

If you want to learn more about how ChatGPT works, read this article.

Do you want to discuss the ideas I outlined? Reach out via Twitter. And if you enjoyed my thoughts, consider sharing the article.

Special thanks to David for helping review this article and providing constructive feedback.

Introducing ResumerX - Elevate your CV using AI

Hans Reinl — Tue, 14 Feb 2023 17:30:07 +0000

Maximize your job search potential with the power of AI! As the buzz surrounding OpenAI's GPT-3 grows, I discovered its amazing potential for enhancing CVs.

With the job market competitive, a standout resume is crucial for landing your next job. That's why I developed ResumerX, a tool leveraging OpenAI's API to elevate your CV and give you an edge in the interview process.

Or, as Chat-GPT puts it: Unlock your full potential with our AI-powered platform and let it edit your CV to new heights. Build on OpenAI's GPT-3 API.

Visit ResumerX and try out for free

And it's easy to use:

✏️ Feed in Your CV Texts: Provide the text from your current CV in ResumerX's platform.
♺ Generate Optimized Texts: Optimize your CV text with AI technology. Make them shorter, extend or change tone.
📝 Use Texts in Your CV: Boost your CV's impact with ResumerX's optimized text. Just copy and paste them into your CV.

My goal with this is to help you improve your CV with no effort, using AI tooling. For recruiters and hiring managers, using AI technology for CV texts will also make it easier to compare CVs - focus is on the content, not form.

Yes, yes, there are lots of tools like this. ResumerX is effort-less, without registration, and for free.

Try out the tool and let me know how it can be improved. Hope it will be helpful to some.

Scrum Refinement: Achieving Excellence in Your Scrum Process

Hans Reinl — Wed, 25 Jan 2023 15:14:08 +0000

Refinements - right?!

Who hasn't experienced some messed up refinements where you thought, "Why am I here? This is a waste of my time". So yeah, doing refinements right, is a hard task.

Let's find out how to do it. But first: Why do we do refinements?
Interestingly, the Scrum Guide does not mention a lot of detail in regard to Refinements. In fact, there are only two sentences:

Product Backlog refinement is the act of breaking down and further defining Product Backlog items into smaller more precise items. This is an ongoing activity to add details, such as a description, order, and size.

That's it. So, it seems like most of what the refinement is, is up for interpretation and discussion.

What we can get from this small description about the Scrum refinement event, or also referred to as backlog grooming, is that it is an ongoing process. This means it occurs throughout the entire product development cycle. It is a collaborative effort between the product owner and the development team, with the goal of ensuring that the product backlog contains items that are clear, concise, and ready for development.

During refinement, the development team and product owner review the items in the product backlog, breaking them down into smaller, more manageable pieces if necessary - it's stated right there in the Scrum Guide. They also prioritize the items in the backlog, based on factors such as business value, risk, and dependencies.

The purpose of refinements is to ensure that the product backlog is always in a state where the next set of items are clearly understood and can be worked on by the development team. This helps to improve the efficiency of the development process, as the team can focus on delivering value to the customer as quickly as possible.

Refinement In Practice

One thing I noticed in the last years working with multiple teams is, that the refinement process is very different from team to team. But why is this?

I personally would argue that most teams are spending too little time on making sure stories are refined enough.
Here I want to share five common mistakes I observed while working with Scrum teams.

5 Common Mistakes with Refinements

There are several common mistakes that teams make when it comes to refinements:

Not setting aside enough time for refinements: Refinements should be a regular, recurring activity, but some teams may not set aside enough time for them, which can lead to a backlog that is poorly defined or out of date. Often times there are arguments like "I need to work on the items in the sprint" or "I don't want to be in meetings all the time".
Not involving the right people: Refinements should involve both the product owner and the development team. If one of these groups is not involved, it can lead to a backlog that does not accurately reflect the priorities and goals of the team. Occasionally, stakeholders should be invited in order to explain their ideas and reasoning behind certain stories.
Not breaking down items into small enough chunks: If items in the backlog are too large or complex, they may be difficult for the development team to understand or work on. It is important to break them down into smaller, more manageable pieces during refinements. This will ensure that team members understand the details of the item.
Not prioritizing items effectively: The product backlog should be prioritized so that the most important and valuable items are at the top. If the priorities are not clear, the development team may end up working on the wrong things. This prioritization also helps teams in pulling items to a sprint if there is room for more.
Not keeping the product backlog up to date: The product backlog should be a living document that is constantly updated and refined. If it is not kept up to date, it may not accurately reflect the current priorities and goals of the team.

By avoiding these common mistakes, teams can ensure that their refinements are effective and add value to the development process.

Next, I want to focus on the primary concern I have observed regarding the effectiveness of refinements: the time invested in implementing them.

Finding Time to Invest in Refinements

The amount of time that a team should spend on refinements will depend on a number of factors, including the size of the product backlog, the complexity of the items in the backlog, and the availability of the product owner and development team.

In general, it is recommended that teams set aside a regular, recurring time for refinements, such as a few hours each week or a half-day each sprint. During this time, the team can review the items in the product backlog and make any necessary updates or changes.

In a previous team I worked with, there were difficulties in completing sprints and the backlog was not well-organized. To address this issue, the product owner and the team decided to prioritize refining the backlog in order to better understand requirements and ensure that the right requirements were being prioritized. This required dedicating a significant amount of time to refinements, such as a daily one-hour refinement session, until the backlog was in a satisfactory state. This approach illustrates that investing heavily in refinements can be necessary to improve delivery and achieve desired results.

It is important to strike a balance when it comes to the amount of time spent on refinements. On the one hand, it is important to ensure that the product backlog is well-maintained and up to date. On the other hand, spending too much time on refinements can be counterproductive, as it can take time away from actual development work.

Ultimately, the key is to find a balance that works for your team and your particular project. It may take some trial and error to determine the right amount of time to spend on refinements, but with practice and experience, you will be able to find the sweet spot that maximizes the value of your refinements while minimizing the time and effort involved.

Reasons Why Not to Do Refinements

There might even be reasons why teams omit refinements all together:

Lack of time: Refinements require a dedicated block of time, and some teams may not have the time or resources to set aside for this activity.
Lack of understanding: Some teams may not fully understand the value of refinements and may not see the need to invest time and effort into this activity.
Poorly defined roles: If the roles and responsibilities of the product owner and development team are not clearly defined, it can be difficult to coordinate and conduct refinements effectively.
Lack of collaboration: Refinements require collaboration between the product owner and development team. If these groups are not working well together, it can be challenging to conduct effective refinements.

None of the given reasons justify not doing refinements. They are merely explanations. You should definitely make sure to fix these issues. Overall, not doing refinements can lead to a product backlog that is poorly defined, out of date, or not aligned with the goals and priorities of the team. This can lead to inefficiencies and delays in the development process.

Conclusion

In conclusion, Scrum refinement is a crucial aspect of maintaining a well-organized product backlog and ensuring the success of any team. It is essential to be aware of the reasons why not to do a refinement and take appropriate measures. The benefits of implementing refinements in your scrum process are undeniable, and although it may not be easy, it is well worth the effort to achieve excellence in your process.

Share your own experiences with refinements, and let's continue to learn and improve together.
Follow me on Twitter and share your story.