詳細検索

Application engineers who remember the balance between profit and challenge

Avatar
by Kiyo Koshiba

Application engineers who remember the balance between profit and challenge
Translated from 日本語 • View original

Hello! This is Koshiba from the Human Resources & Branding Project!

**Let's interview Colorkrew's engineers, Part 12! **

This time, we talked to Mr. Yamana, who recently pickled kimchi. In addition to making kimchi, Mr. Yamana has antennas in many ways, such as making a hygrometer with a raspberry pie and reading the Nikkei news without turning it on. I was mainly involved in the development of core systems for payment agencies, but recently I am an application engineer who has expanded my scope to cloud infrastructure and data science.

The story of making a hygrometer with a Raspberry Pi

[caption id="" align="alignnone" width="800"]

Zoomでインタビュー

Since I was in charge of the Odakyu project, I appeared with Hakone soba in hand. [/caption]

Koshiba: Recently, I often see posts about Raspberry Pi on Goalous (internal SNS), but that's a hygrometer, right? What made you start?

Yamana: Previous Award (an award given to those who have made a significant contribution once every six months. If you are selected, you will receive additional bonuses. We received the Team Gold Award as a billing and settlement team. I used that money to buy a dehumidifier, but I wanted to know how effective dehumidification is.

Koshiba: Didn't you buy a hygrometer?

Yamana: I wanted to know how it changes over time, so I wanted something that could keep records. But I only had something for research that cost tens of thousands of yen, so I thought it would be possible to do a raspberry pie, so I looked it up and found an article on Qiita that was just right. So I decided to make it.

Koshiba: I see.

Yamana: Raspberry pies don't cost 10,000 yen even if they have high specs, and you can customize them to your liking. You can connect it to the cloud or view information in BI.

Koshiba: You can use Colorkrew's technology in various ways.

Yamana: However, I bought the dehumidifier around July, and I actually started making it in Raspberry Pie in September, and it was finished in October, so the period of high humidity was over. That's why I ended up with self-satisfaction that I just made. Laughter

Koshiba: It's not about dehumidification, it's about entering the humidification season. lol But you're trying to expand it to the company, right? Is it for office improvement?

[caption id="" align="alignnone" width="599"]

ラズベリーパイで湿度計 Share progress on internal SNS. [/caption]

Yamana: At first, I started for fun, but it seems that temperature and humidity are important in terms of hygiene. This is also the case, and < to improve the way we work is a href="https://mamoru-secure.com/?source=pay&utm_source=engineer_vol12&utm_medium=referral&utm_campaign=owned" data-md-link>Mamoru concept, so I'm doing it now, thinking that there is a possibility of making it optional.

Koshiba: That's interesting. It is said that if it is dry, it is easier for the virus to be infected. When making an office hygrometer, I would like you to know not only the relative humidity but also the absolute humidity!

Yamana: Oh, that's a good idea! I will probably be able to get absolute humidity, so I'll check it out.

Koshiba: I did it! Being able to add functions in this way and create something with your own hands is certainly more fun than just buying a hygrometer!

Cultural Differences by Project

Koshiba: It's time to leave Raspberry Pie, and next time can you tell us about your work?

Yamana: Yes, right now, about 50% of the billing and payment work is done, about 40% of the cloud MSP work, and the remaining 10% are AI-related jobs. I would like to increase the proportion of AI in the future.

Koshiba: What kind of work do you do in billing and settlement?

Yamana: I am doing two main things now. One is daily operation. In order to respond to customer requests, I do PM-like things, such as assigning work within the company. The other is to upgrade the PHP version, and in this project, I am also working on it.

Koshiba: Mr. Yamana = I have an image of billing and payment, but now I am working as a cloud MSP for 40%.

Yamana: In the cloud, we build data processing pipelines with Azure Data Factory. When I was developing a portal site with Mr. Hotta, I was invited to say, "Are you interested in data processing?"

[caption id="" align="alignnone" width="800"]

エンジニアに聞いてみたブログ

Click here for an interview with Mr. Hotta who invited Mr. Yamana. [/caption]

Koshiba: That's right.

Yamana: At that time, I didn't know anything about Azure at all, and there was a time when it was quite difficult because I was stuck, but I didn't know about it, so I had to look it up! I have no choice but to ask! I managed to catch up.

Koshiba: After all, there is such a difficulty when you take on the challenge of a different field.

Yamana: That's right. Now we are a team of about 5 people, and those who can do it take tasks and develop them. I could take on challenges that suited my height, so it was perfect for me now. Of course, since I raised my hand, I have to take responsibility for it and finish it.

Koshiba: Did you feel a difference between billing and cloud MSPs when you were involved in both billing and cloud MSPs?

Yamana: Culture varies depending on the project. What is normal for billing and payment teams is not for cloud MSP teams.

Koshiba: For example, what kind of things can you give?

Yamana: The billing and payment team can't make mistakes in many ways. The pressure is enormous because it is directly related to customer sales. We can't make even the slightest release, so both the customer and we are very cautious. We always review it and do not release it if there is even the slightest concern, and the flow is established when a problem occurs. Anyway, I'm very cautious.

Yamana: Of course, I am cautious because I work with a variety of customers in my cloud MSP work. There are also tests. But I think it's more flexible than the billing and payment method.

Koshiba: It's not about which one is good or bad, but there are characteristics and culture that suit the business.

Are you making meat and potatoes? Is it curry?

Koshiba: Mr. Yamana personally, what kind of policy do you have for development? Is there anything you value to you?

Yamana: I'm a little particular about how to put comments.

Koshiba: What do you mean?

Yamana: In the end, it's up to you to decide who the comments are useful for. Even the code I wrote was forgotten after a month.

Koshiba: That's right.

Yamana: If there are no comments, it will be difficult to remember, and if you are not good at it, there is a possibility that you will write strange things and cause problems. That's why I think comments are important. So, in my case, I try to write about what I am writing for, not what I am doing.

Koshiba: Do you mean to clarify the goal?

Yamana: Yes. If you know what the final form of treatment is, it is easier to understand that it is better to do this or that.

Koshiba: Hmm, I don't seem to understand...

Yamana: If you don't know if what you're making while being instructed by the side is meat and potatoes or curry, you don't feel like you can make delicious foodThat's right.

Koshiba: I see! I really understood!

Balancing Earning and Challenging

Koshiba: What kind of challenges do you want to take on in the future?

Yamana: As I mentioned earlier, I would like to increase the proportion of AI work.

Koshiba: Even if we increase it, it seems difficult because we don't have enough projects yet.

Yamana: I jump on the information dissemination of the sales person in charge of data science projects, and I am allowed to accompany them.

Koshiba: It's a steady effort.

Yamana: What I found out after meeting and talking with various customers is that core AI technology tends to be developed in-house without releasing it to the outside world because it is full of confidential information. In such a situation, I think that in order to break into AI as Colorkrew, we will cleanse the data that we did not intend to use for AI so that it is easy to put it into AI. How to turn it into a business is an issue.

Koshiba: Following sales and discovering business is quite ... no, it's quite aggressive.

Yamana: I had never interacted with customers until I joined Colorkrew, so I was nervous about answering a phone call. That's no longer the case now. Laughter

Koshiba: That's a big change!

Yamana: I became aware of money after joining Colorkrew. It's a flat organization, so if you raise your hand, you can do what you want. But in order to continue doing what you want to do, you can't continue without making a profit. It doesn't matter if it's an engineer or a salesperson, and the problem of numbers is not just for PL, but for everyone to be concerned about. That kind of awareness became ingrained in me after I came to Colorkrew.

[caption id="" align="alignnone" width="800"]

エンジニアに聞いてみたブログ

It is important to balance the challenge of new things and earning profits. [/caption]

Yamana: Going back to AI, fortunately, we have products such as Mamoru and Goalous within our company, so we are working to involve AI in themI plan to do it too.

Koshiba: There is a lot of data.

Yamana: For example, Goalous has been posting more recently due to the influence of remote work. I think it would be useful for people to recommend that the posts they bring to the top change depending on the person, for example.

Koshiba: That's good! I feel that it will be very convenient if the areas you want to do are added to the product, so I would like you to come up with a lot of ideas and implement them to evolve into a good product.

Yamana: Yes, I will do my best!

Koshiba: It is quite a difficult problem to establish the area you want to do as a business, but I want to make it happen somehow. Thank you very much for talking to us today!

Related Articles