My Data Guest — An Interview with Ángel Molina

This is the interview translated by the KNIME team that Roberto Daniele Cadili did with me in My Data Guest

It was my pleasure to recently interview Ángel Molina as part of the My Data Guest interview series. Ángel is many things. He is a management controller, data analyst, and a data science influencer on several social network platforms. In just a year and a half, Ángel seamlessly took on the role of a data science reference for the Spanish-speaking KNIME community alongside his administrative duties. Moreover, he earned the title of COTM in September 2023, and again in January 2024.

With Ángel we talked about his journey from an administrative assistant to a data expert, his roles as a blogger and data science influencer and lastly, since he runs a dedicated newsletter, his take on the ever more important link between data science democratization and AI.

The original interview was in Spanish, here you can read the English version.

***

Roberto: Tell us more about your professional self and the role you feel most connected to.

Ángel: I have many professional selves but the roles I identify the most with are those of a management controller and data analyst. My main occupation is helping manage a medium-sized company located in Córdoba (Spain) as a management controller. When my working day is done, I dedicate myself to something I love, which is sharing content with the Spanish-speaking KNIME community. I write blog posts on different topics in data analytics and I’m very active on the KNIME Forum.

Roberto: What are your daily tasks?

Ángel: In the morning, I help manage and organize different areas of a company, such as production, finance, sales, analytics and accounting. In the afternoon, I unleash creativity and learning by writing a blog post or producing educational content. It really depends on how inspired I feel.

Roberto: Is there room for automation in management control?

Ángel: Definitely! But not only in my company, in any company. Even in very small ones, there is always room for automation. One example I could think of off the top of my head is automating bank reconciliation. This was also the focus of a recent blog post of mine. Bank reconciliation is usually a very tedious and error prone process, which makes it the perfect use case for automation in management control.

Zooming out from administrative tasks, a lot of work in other areas entail performing repetitive tasks, so automation can really be leveraged anywhere.

Roberto: Do you feel there is some resistance to letting go of traditional tools (for example, spreadsheets) within your professional circle?

Ángel: Absolutely! Imagine introducing a new tool to professionals who have been using the same set of tools for ages…It’s really hard for them to see the value of learning a new tool.

Apart from resistance, the underlying cause is always a lack of knowledge. Many people often do not know about tools like KNIME. One of the reasons why I recently decided to be active on platforms like TikTok and Instagram is to show spreadsheet users, who may still be using Excel, how much easier it is to automate repetitive tasks with KNIME.

Roberto: You are proof that data analytics can be incorporated in any task and processes can be improved. How did you get into the world of data analytics?

Ángel: First and foremost, if I could make it, anyone can. I began my journey with advanced functionality in Excel. Then, I ventured into Power Query, Power Pivot, and Power BI, and dabbled in Python. While delving into Python, –don’t ask me how or when– I accidentally stumbled upon KNIME.

Roberto: How was the encounter with KNIME? What was your impression back then?

Ángel: I was searching for new analytical tools, and I ended up attending the 2021 KNIME Summit, which was held online because of Covid-19. Back then, I had no idea about the software but what positively surprised me was the outstanding community behind it so I said to myself: “OK, I’m staying and I want to learn more about it”.

Roberto: What was your “wow moment”? The time when you realized that a low code tool could actually solve problems that you didn’t think possible.

Ángel: Python is great but to unleash its potential you have to learn a programming language and play around with libraries and environments. This usually takes a lot of time and expertise. On the other hand, KNIME has the same potential but it’s much easier to learn and use thanks to its visual interface. Moreover, if you are still missing something from Python and you want to incorporate a code snippet, you can do it by just using the Python Script node.

If that wasn’t enough, with the latest software release of KNIME Analytics Platform (v5.2), you even have access to K-AI, KNIME’s free AI assistant that helps you build workflows, code in Python or answer questions about KNIME nodes.

Roberto: Have you had the opportunity to experiment with K-AI?

Ángel: Yes, and it’s truly fantastic. I think it goes in the direction of democratizing data science, enabling people who do not have a technical background to take their first steps in the world of analytics and make sense of data. You really have no more excuses at this point. Give K-AI a try –it doesn’t get any easier than that!

Roberto: Have you incorporated KNIME into your job? What features of the tool do you use the most?

Ángel: In my day-to-day work, I focus mostly on ETL (Extract, Transform, Load) operations. With KNIME, it is very easy to automate any sort of ETL process –something that would be much harder to do (if at all possible) in Excel. I haven’t had the chance yet to delve deeply into the new reporting nodes but they look very powerful and promising. I can totally see myself using them in my daily tasks to create static PDF reports.

Roberto: As I’ve mentioned, you are also a data analyst and consultant. How did that career change or expansion come about? Could we say that KNIME has opened up more paths for you?

Ángel: It really did. KNIME has opened up more professional opportunities and boosted my career prospects. The fact that we’re running this interview is just another example. Once I became a KNIME user and an active community member, people and companies started asking me if I could help them solve problems with KNIME, so that’s how it started.

Roberto: Which functionality of the software do you consider has facilitated you the most in this career transition?

Ángel: All the functionalities were important. ETL capabilities are crucial for a data-handling software like KNIME, but I believe that what truly sets KNIME apart are the extensive resources for data modeling and machine learning. In the past, I carried out some machine learning projects in Python but it was often quite annoying and time-consuming to deal with dependencies, installations, environments. On the other hand, KNIME managed to turn something quite complicated into a very easy and smooth process –a blessing for people like me who do not have a background in Math or Computer Science. The AutoML component, which can be used to automatically train several supervised models, is just one of the many great examples.

Roberto: What are your 3 favorite nodes?

Ángel: Due to my job, I usually carry out many data wrangling operations. Hence, my go-to nodes are the ones for ETL:

• GroupBy
• Joiner
• Row Filter

You can easily find these nodes in all my workflows and probably in yours too, Roberto! I also want to name my favorite extensions:

• The Geospatial Analytics extension, co-developed with the Center for Geographic Analysis at Harvard University. It’s probably one of my favorite topics to write about in my data stories, since it’s very visual and catchy.
• The KNIME AI Extension and in particular the connectivity with GPT4All, which allows you to work locally with LLMs like Llama 2.
• The Web Interaction extension with Selenium, which is truly wonderful, as it makes data scraping from web pages incredibly easy.

Roberto: You write in English on Medium, and you publish not one but two weekly newsletters on LinkedIn in Spanish. For that constant and inspiring work, you won the COTM in September 2023. Did you feel that there was a need for more content in Spanish?

Ángel: I started off just by writing blog posts where I shared my solutions with KNIME. I saw that as a way of giving back to the community for all the help I received when I took my first steps with the tool. At first, I was writing only in English because that was what everyone else was doing, but I later realized that nobody was writing in Spanish so I tried that out and it worked very well. The response of the Spanish-speaking community has been amazing!

Roberto: How do you choose the content of your blogs? Who is your audience?

Ángel: First of all, I always choose a topic I like. If I don’t enjoy what I’m writing about, it’s going to be a waste of time and my readers won’t like the content either. Secondly, I choose something that I think is of interest to the readers. For example, some new node or extension that I think is underutilized (e.g., the AutoLM component).

My audience is usually data analysts or data scientists from medium or large companies in Spain and Latin America. I’m currently trying to change that, as I would like to gain more followers who have zero knowledge of data science. The goal is to help absolute beginners start their data analytics journey.

Roberto: Some of your most appreciated blogs are about AI and LLMs. Beyond the hype surrounding this topic, why do you think they are so popular?

Ángel: At first glance, LLMs can be intimidating because of their complexity but my goal is to write about them in such a way that even a six-year-old can understand what I’m saying. I believe that the key is to use a clear and simple language, complementing the text with images or videos. When non-technical people read blog posts about AI in newspapers or magazines, there are always concepts or words that they do not understand. I believe it is valuable to inform them about how the world is changing and, to achieve this, you need to adopt a simple and plain language.

Roberto: Writing about complicated things in a simple manner is indeed a very hard task. Talking about writing, one of your data stories about web scraping with the Web Interaction extension was featured in the annual list of the top 10 blogs in the KNIME community in 2023. Tell us more about it.

Ángel: I had already done some web scraping in Python, because it can be a real asset for any company. For example, think about the possibility of automatically collecting and then checking in real time the prices of your competitors vs yours.

One day, I was browsing the KNIME Community Hub and came across the Web Interaction Extension. I was hooked immediately and, needless to say, it turned out to be one of those nights where you go to bed at 2 AM. As I was building my first workflow with those nodes, I already started thinking about the blog post and how KNIME helped me achieve something very complicated in an intuitive way.

Roberto: Your LinkedIn profile has reached more than 11k followers in a very short time. This means that your content is good and users appreciate it. Would you like to make a living as a data science influencer with KNIME?

Ángel: I have a really large family so some of the 11k followers must be my cousins and aunts. Jokes aside, I’m very happy about that! However, the word “influencer” feels a bit too much, to be honest. I simply try to share and write about topics in data science that I find interesting and, hopefully, valuable for others too. I will certainly continue to do it in 2024, and who knows what the future will look like?

Roberto: Have you ever thought about creating your own KNIME course?

Ángel: I did. I recently ran a survey on LinkedIn and I received positive feedback from about 50 people who would be interested in taking part. The course would obviously be in Spanish.

Roberto: With the light-fast evolution and spread of AI that we have witnessed in the last years, I cannot help but ask you the million-dollar question. How do you think data science will evolve in the coming years?

Ángel: That is, indeed, a hard question to answer. I think there will be areas of data science that will be accessible with less effort than before thanks to the introduction of AI assistants like the ones developed for Python scripting or K-AI for KNIME. However, this is only in the short term. The medium-long term trajectory is really an unforeseeable scenario. I even thought of writing a book about AI but then I realized that by the time the book is ready for publishing, I’ll have to re-write it again or write a whole different one.

Roberto: Before we say goodbye, is there any exclusive news on your plans for 2024 for the audience of My Data Guest?

Ángel: I wrote a book that will come out soon and it collects all the blog posts I have published about KNIME. Someone special from KNIME will write the preface and it will be available for free to all my subscribers on social media. And who knows? Maybe I’ll also start my own YouTube channel.

Roberto: How can people stay in touch with you?

Ángel: You can reach out to me and follow my work on social media: LinkedIn, Twitter, Threads, Instagram, TikTok, KNIME Forum and Medium. Happy to hear from you!

Watch the original interview in Spanish with Ángel Molina on YouTube:

My Data Guest — Ep. 29 — Ángel Molina