Machine Learning Engineer | Global Manufacturing Business

【About the Company】

【Unleashing the potential of the manufacturing industry】

The company will create a society in which all people involved in manufacturing can maximize their inherent power.
To achieve this goal, they will create a "new mechanism" that will change the common sense of industry.

Busy with estimating and administrative tasks, lacking sales skills, and lacking information and networks.
By untying these shackles, the potential of each company can be unleashed.
From small factories in town, to large manufacturers with a long history, to start-ups in their early years, all manufacturing companies will shine by leveraging their strengths to create new value.
They will continue to take on the challenge to open up such a future.

◆Flat organization
Regardless of your position, team, previous experience, gender, age, etc., you can express your opinions and immediately incorporate what is good.
◆Excellent members
The team is made up of members who have been active in leading companies. You will have an environment where there is a growth and stimulation every day.
◆Engineer Driven
Since they are operating in a field where there are no precedents yet in the world, technical skills are extremely important, and engineers play a leading role.

--------------------------------------
◎A service that uses automated quotation technology to connect custom-orderers with processing companies.

The manufacturing industry is Japan's key industry, boasting a gross domestic product of 180 trillion yen.
In fact, about 120 trillion yen of that is accounted for by the cost of parts procurement.
Despite this large percentage, there has been no major innovation in the procurement field for over 100 years.
In particular, parts procurement for the high-mix low-volume production industry, which accounts for about one-third of the total, has been facing various social issues on both the ordering and receiving sides, such as the time and effort required for ordering and quotation, procurement costs, and the high deficit ratio on the production side.

They believed that by solving these problems, manufacturers not only in Japan but also around the world would be able to focus on higher value-added work and maximize the potential of the manufacturing industry as a whole.
So, the company developed the world's first service that uses automated quotation technology to connect custom-orderers with processing companies.

【Job Description】 

【Recruitment Background】
With the mission of "Unleashing the Potential of the Manufacturing Industry," the company is developing two businesses: CADDi MANUFACTURING, a one-stop partner for procurement and manufacturing, and CADDi DRAWER, a product that converts data in the supply chain into assets.
CADDi DRAWER, launched in 2022, uses machine learning and other technologies to structure drawing data, which is said to be the most important data in the manufacturing industry, and link it to various information, making it possible to use it as an information asset. It is already being used by customers ranging from major domestic manufacturers to processing companies, and sales have begun overseas (in the U.S. and Vietnam) from 2023.

In the future, they aim to realize overall optimization that transcends divisions and companies by using technology to reproduce and consolidate manufacturing industry knowledge beyond drawings. Beyond that, they hope to transform the work styles and cultures of their customers, leading to reform of the manufacturing industry as a whole. So, the company is looking for people who can work together to create such a world.

Your Role
As a Machine Learning Engineer, you will develop models in machine learning and data science, and build, maintain, and operate the infrastructure that enables the continuous delivery of these models to our services. You will be expected to develop highly accurate modeling techniques that can provide value to the company's products by utilizing the company's data, and to develop stable systems in a team environment.

【Examples of Work】
Construction of image recognition system for drawings
You will analyze the images of our partners' drawings and develop technology to extract information on the drawings.
*Actual duties are not limited to the above. Your responsibilities after employment will be determined based on your skills, expertise, experience, and other factors.

Development and deployment of batch processing and API to extract information on drawings
Feature extraction from images, and construction, maintenance, and operation of a similar image retrieval system using these features
Build classification models of drawings using Deep Learning, and create annotation systems
Creating demos and reports of the created image recognition models and explaining the technology internally and externally
Hearings from internal and external parties on issues related to drawing information and definition of tasks that can fulfill the requirements
Experimentation, analysis, and visualization to ensure high model accuracy

Experience gained
Experience working with highly motivated team members on challenging technical issues
Experience working with members who have expertise in a wide range of areas, including software as well as machine learning
Experience in solving problems by taking into account how to develop business value from the technology
Close proximity to MLOps and product management members, and the ability to expand the scope of your work depending on your will.

About the Team
The CADDi DRAWER Group currently has approximately 50 engineers involved in development.
The team structure is based on Team Topologies, and the organization has the following characteristics

Team Structure: Divided into 9 teams (3-6 members per team) including Functional Development, ML/MLOps, Data Pipeline Development, Enabling (QA/SRE), etc.
Multinational organization: 20% of development members are from overseas (Asia, Europe, North America, etc.). Some teams communicate mainly in English, and important meetings are held in both English and Japanese and the company is trying to create an organization in which multinational members can play an active role.

【Working conditions & treatment】

Flextime System

・Saturday/Sunday/National Holiday
・Special Paid Leave
・Congratulations & Condolence Leave)
・Bereavement Leave
・Refresh Leave
・Summer Holiday

・Full Social Insurance
・Commuting Allowance
・Child Allowance
・Medical Checkup
・Training/Self Development Allowance
・Commuting allowance (up to 30,000 yen)
・Vacation (summer vacation, year-end and New Year's vacation, refreshment vacation, bereavement vacation, etc.)
・Subsidies (moving subsidies, child allowances, marriage congratulation money, etc.)
・Medical checkups
・Office medicine
・Office convenience store
・Learning support (book purchase system, language learning support, manufacturing experience, external training support, etc.)
・Company-wide awards
・Club activities
・Engineers can apply for a PC and display with their desired specifications.
※The maximum amount is 400,000 yen, within which you can also purchase accessories for the PC.
※The PC replacement cycle should be at least two years.

【Required】
Basic knowledge of algorithms related to machine learning, statistics, linear algebra, and computer science
Experience working with machine learning to solve business problems
Experience in improving the accuracy of machine learning and statistics models
Experience developing and operating APIs related to web services using Python, Rust, etc.
Experience working with cloud services such as Google Cloud and AWS
Basic knowledge of container technologies such as Docker
Experience in team development and operations using Git and CI/CD
Fluent business communication in Japanese (N1 or above)

【Preferred】
Experience with GPU-based data processing (CUDA, OpenCL, cudf, CuPy, etc.)
Experience developing machine learning pipelines using Vertex AI Pipeline, kubeflow, Apache Beam, Spark, etc.
Experience continuously improving and delivering machine learning and data science models
Experience implementing data quality measures to improve machine learning models in a Data-centric manner
Experience winning multiple prizes in data analysis competitions such as Kaggle
Experience writing papers in leading journals in the fields of machine learning and data science
Experience in applying numerical optimization methods to business problems
Experience in web service development related to front-end and back-end, and experience in development and operation related to distributed processing

【Ideal Applicants】
Empathy with the company's mission "Unleashing the potential of the manufacturing industry
Eager to learn and take on new technologies and things they have no experience with
Willing to catch up on relevant technologies required for ML/MLOps
Able to face essential issues and take actions to solve them with a sense of ownership
Able to work through positive attitude and constructive discussions in a fast-changing and uncertain environment
Able to communicate and discuss with others in a respectful manner, taking into consideration their context and resolution