Before you go, check out these stories!

Hackernoon logoWorking at a FoodTech Startup: A Data Scientist's Insights by@databug

Working at a FoodTech Startup: A Data Scientist's Insights

Author profile picture


Data Scientist for Food Tech. Based in South Korea

Since 2018, I have been working for a plant-based alternative startup [더플랜잇, The Plant Eat] as a data scientist. Here is what I learned while working on the food data team.

Reminder: This is my own opinion, it doesn’t represent the company’s ideas and views regarding this topic.

Here are the points that I learned during my journey:

1. Food can be modular but very dependent on mixture of compounds

A single ingredient may influence your product mix regardless of the ratio of your food product. (i.e. single drop of sulfur-based ingredient may create an animal-based food like taste/aroma)

2. Food preferences may be influenced by compounds likeness or distance between the two foods.

Let me quote an article titled “Flavor network and the principles of food pairing” that explains this point

North American and Western European cuisines exhibit a statistically significant tendency towards recipes whose ingredients share flavor compounds. By contrast, East Asian and Southern European cuisines avoid recipes whose ingredients share flavor compounds

3. Food Data needs a lot cleaning, I mean a lot of cleaning

I can’t show the specifics but the food industry does have diverse weighing systems, processes in getting numerical data, and other diverse data from the lab that is traditionally saved in numerous types of file format that depends on what kind of lab machine you are using. (i.e. BRIX, to measure the amt. of sucrose, a type of sugar)

4. Choose what kind of food characteristics you are going to concentrate on. Is it function?, taste?, nutritional?, or just the price?

Although not impossible, it would be wise to concentrate on one factor at a time since each factor have a different indicator and unique condition that is required by the Food Administration or your target consumer.

5. Production is king. If you can’t scale it up then it is not feasible.

Yes, you got all your product ready but if the factory doesn’t have the machinery or the skill set to create a stable product… it will cost you a lot of money and time to make it work. Always research on the factories that you are going to work with for production.

6. Patch updates isn’t isolated in the software space but very relevant in the food science space as well.

R & D based Food Tech companies need the software-like “Patch Update” Thinking. Rapid updates and consumer-centric products will be easier through this method. If you have the Software “Update Patch” in Tech Space, Food Tech has a “Formulation Patch”. Formulation Patch meaning that food products would have version updates very quarter with small but significant changes.
theplanteat is taking this approach as well creating internal software to keep up with the fast feedback we are getting in the market today.

These are the general insights I learned so far while working for a food tech startup. Some of the insight is not included due to the sensitivity of the data.

Thanks for Reading!

Previously published at


Join Hacker Noon

Create your free account to unlock your custom reading experience.