Brazil is a huge country for coffee. Not just as the largest producing country, but as a staple for many a roastery. The last decade has seen a huge change in appreciation for this vast origin, not least with the growth in regional profiles and experimental processing. But who better to ask about how best to roast it than those who are in the middle of it? We sat down with SMC’s Master Roaster, Raul Paulino, to chat about all things coffee roasting.
DRW: Tell us a little bit about yourself. What is your role? How many roasts do you do a day? What roaster(s) do you use?
Raul: My name is Raul Felipe Paulino de Souza. I’m 30 years old and hold a degree in Agribusiness Management. I am also a licensed Q Grader and professionally certified in roasting by the SCA. I have competed in two Brazilian roasting championships sanctioned by World Coffee Events (WCE), finishing as runner-up in 2022 and in fifth place in 2025.
I have worked in quality control at SMC Specialty Coffees for 12 years and am responsible for the roasting department. I began specializing in roasting in 2018, and since then, it has become a true passion and area of intense focus for me. I’m constantly eager to learn something new and gain a deeper understanding of Brazilian specialty coffees’ potential.
During the busiest period of the harvest season, we roast around 250 samples per day. We receive a wide range of coffees from different regions, varieties and drying processes throughout the season. We therefore need to adapt constantly, drawing on both our technical knowledge and the intuition we have developed over the years. We roast exclusively for sensory evaluation, so our focus is on developing roast profiles that allow each coffee to express attributes such as sweetness, acidity and body in a balanced way, without masking its natural characteristics.
Our lab is equipped with three types of sample roasters: two Probat machines, being one with four drums and the other with two and both with a capacity of 100 grams per drum and gas heating; a gas-powered Probatino with a 1.2 kg capacity; and a Kaleido M2 with a 400-gram capacity and infrared heating.
DRW: What is your methodology when approaching a roasting day?
Raul: Before defining a roast profile, we conduct several analyses to better understand the coffee:
• Check its moisture content using a Gehaka G200;
• Measure its water activity (aW) using a Coffmeter W1;
• Analyze its density;
• Assess the drying process based on the coffee’s physical appearance, seeking to determine whether it’s a natural, pulped natural, or has undergone an induced fermentation process;
• Check on the ambient temperature;
• Consider the coffee-growing region and our previous experience with that type of coffee.
Based on these analyses, I preheat the roasters and determine the charge temperature, flame or power setting, and airflow adjustments.
DRW: How do you deal with variation between moisture, bean size, etc. or does that not matter?
Raul: At SMC, coffees with moisture levels that are either too high or too low are rejected before roasting, so there is no need to roast them, which makes our work considerably easier. When roasting coffees with smaller screen sizes, I generally begin with a lower charge temperature and apply less flame or power throughout the roast. For larger screen sizes, I use a higher charge temperature and greater energy input.
For denser coffees, I increase the charge temperature and apply more flame or power to prevent the coffee from getting “cooked.” For less dense coffees, I use a lower charge temperature and less flame or power, preventing the roast from progressing too quickly or developing defects such as tipping and scorching.
DRW: What do you think works best in roasting Brazilian coffee?
Raul: It’s difficult to explain because it does rely heavily on intuition. Theory is extremely important when comes to decision-making during the roasting. However, after completing many roasts and learning to understand how coffee behaves, the process begins to feel more intuitive.
Coffee never lies. If you do something unsuitable for a particular coffee, it will reveal itself in the cup. After several years of roasting, I have learned that there are no absolute rules and no universal right or wrong approach. Ultimately, what truly matters is how the coffee performs in the cup. Nevertheless, it is essential to have clearly defined procedures and a thorough understanding of the roaster you are using. This provides reliable parameters and enables you to plan and execute the roast properly.
I have noticed that some coffees with more pronounced fruity and floral characteristics respond well to shorter development time ratios, around 12%, while coffees with greater sweetness potential may perform better with ratios between 15% and 18%. In both cases, specifically for sample roasting, I generally work with a total roast time of between eight and ten minutes.
For these sample roasts, I prefer to use no gas – or keep the power at zero – until close to the turning point, preventing the coffee from being exposed to excessive energy during the initial stage. Throughout the roast, I aim to maintain a consistently declining Rate of Rise.
Roasting is a science as much as it’s an art. You need to understand what the coffee is asking for and make the right adjustments. Brazil has an immense range of coffee varieties, along with a growing number of new drying and fermentation processes. This is fantastic for our origin, but it’s also challenging. As we receive many different coffees, we must adapt and learn to understand each one – and that is precisely what motivates me.
Throughout the harvest season, we welcome many clients, including groups from DRW. I make the most of these opportunities to speak with roasting professionals, many of whom have far more experience than I do. I seek to understand their perspectives and absorb as much knowledge as possible so that I can continue improving. This kind of interaction really valuable for me.
DRW: How does your approach differ from pulped naturals to naturals?
Raul: In my experience, pulped naturals often respond more resiliently during roasting than natural coffees and require greater energy input throughout the process. However, I don’t attribute this difference exclusively to the processing method. Factors such as moisture content, water activity, bean integrity, and storage conditions can have a much greater influence on how a coffee behaves during roasting. I therefore assess all these characteristics together before defining the roast profile.
DRW: Have the new processing methods changed your approach to roasting?
Raul: Absolutely. New processing methods are emerging all the time, and they can affect how a roast should be developed. Depending on the fermentation process and drying conditions, some coffees may have lower density and greater sensitivity to heat. In these cases, I tend to use lower temperatures and reduce the energy input.
Certain processes can also affect the final color of the roasted beans. Color plays a significant role in roast evaluation, so when different processing methods alter the appearance of the roasted coffee, visually assessing the degree of roast development becomes particularly challenging.
DRW: Do you have regions where you change your approach, or do you treat them all the same?
Raul: I don’t approach coffees from every region in the same way, but neither do I change a roast plan simply because a coffee comes from “Sul de Minas” or the “Cerrado.”
For me, the region is a starting point for anticipating how the coffee might behave. I then make adjustments primarily based on density, moisture content, screen size, and processing method. I see the region more as a working hypothesis than a rule.
On the other hand, when these coffees are processed properly, some can exhibit remarkable aromatic complexity. If exposed to excessive energy or an overly long development time, they may lose precisely the qualities that make them special.
DRW: Do you find any differences between roasters?
Raul: Yes, significantly – especially when the roasters use different heating systems. I notice the greatest differences in the aromas perceived during roasting.
With gas-fired roasters, the aromas are much more distinct during the roast than with infrared roasters, for example. In the cup, however, it’s possible to achieve very similar sensory results.
We conducted tests by roasting the same coffees under the same conditions in both the Probatino and the Kaleido, and achieved very similar sensory results. Some roasters respond more quickly to adjustments than others, while the balance between conduction, convection, and radiation varies according to the design and construction of each machine. For this reason, I believe that thoroughly understanding how the equipment behaves is just as important as understanding the coffee we are roasting.
DRW: We agree! Thank you so much for your insights.
One of the key reasons for visiting origin we have found over the years is to have space to have these conversations. Not just with the producers but other actors in the industry where huge amounts of passion and knowledge is shared. You can read about our last trip to Brazil where Jack visited SMC and his thoughts on what it means in our trip report.
