Data Scientist (Specialist)

Summary

This position requires utilizing machine learning and statistical models to extract valuable insights from customer information in order to solve customer issues.

We are particularly interested in engineers with expertise in advanced model development, feature engineering, and optimization methods, with a proven track record in Kaggle competitions and research-level machine learning algorithm development experience.

Main Responsibilities

1. Design, implementation, accuracy evaluation, and improvement of machine learning and deep learning models
2. Pre-processing and feature engineering of large-scale data
3. Analysis and predictive modeling of time-series and unstructured data
4. Producing technical reports based on analysis results
5. Automation of analysis pipelines and development of MLOps environment

Case Study

We started in earnest in December 2019 and have worked on the following projects.

[Demand Forecasting]

1. Demand forecasting in the food distribution industry
2. Demand forecasting for parts in the automotive industry
3. Power generation forecasting in the renewable energy industry
4. Commodity price forecasting in the financial industry

[Mathematical optimization]

1. Mathematical optimization for minimizing inventory and shortages in the food distribution industry
2. Mathematical optimization of warehouse inventory in the automobile industry
3. Optimization of steel sheet cutting operations and delivery of components in the steel industry・Combinatorial optimization to improve delivery efficiency in the distribution industry
4. Truck layout optimization to improve production per cost in the mining industry

[Personalization]

1. Churn forecasting for subscription business in the electric power industry
2. Customer analysis and marketing support in the automotive industry
3. Default forecasting in the banking and finance industry

You can be involved in a wide range of global industries (industrial plants, logistics, energy, retail, food, healthcare, urban development, mobility, power, chemicals, metal resources, etc.) that Mitsubishi Corporation handles. In addition to development, you will also be able to take the initiative in creating your own business by entering into projects from the planning stage of new business with MC's staff.

Not only PoC, but data scientists and software engineers work together to develop models created by data scientists, and work all the way through to the provision of production systems and AI SaaS.

Tech Stacks:

1. Machine learning : PyTorch, LightGBM, Optuna, etc.
2. Reinforcement Learning : Stable Baselines
3. Machine Learning Efficiency : Luigi, Kubeflow Pipelines
4. Mathematical Optimization : OR Tools, Gurobi, etc.
5. Google Cloud Infrastructure: Google Compute Engine, Google Cloud Storage, BigQuery, Vertex AI, etc.
6. Collaboration tools: Slack, Github, Confluence, etc.

About our employees

We have about 80 employees, half of whom are software engineers and data scientists. Many of our software engineers are former employees of well-known tech companies. Our data scientists include top rankers in the Kaggle programming contest, medalists in the International Olympiad of Informatics, and PhD holders.

Required Skills:

1. Experience in data analysis or machine learning model development using Python
2. Basic knowledge of SQL and databases
3. Experience using version control tools such as Git
4. Basic knowledge of statistics
5. Business level Japanese or English skills

Preferred Skills:

1. Top rankings in Kaggle competitions or equivalent
2. High performance in programming contests
3. Experience in publishing papers in prestigious international conferences or journals in the field of machine learning
4. Experience contributing to OSS
5. Experience in solving problems by mathematical optimization
6. Experience developing machine learning models on cloud services such as Google Cloud
7. Experience in team development or problem solving
8. Familiarity with reinforcement learning, generative AI, or time series prediction