Job Requirements
Less than a year of experience
This job post is managed by
choon.mui.lee
Last active 2 years ago
Skills
Job description for Data Scientist Intern (Forecasting) at Stemly
We are looking for a Data Scientist Intern to join our team. You should have a solid background in Programming and passionate about data science, machine learning and problem solving. You would also need to have a basic knowledge about machine learning such as training and evaluating machine learning models.
Project Scope
This project aims to design and implement a prototype for an automatic feature extraction module for time series data. We want to be able to predict the type of features that will be useful for a particular problem, that is also explainable in a business context. This feature extraction process will be important for developing meta-learning techniques for forecasting, which will also automates the model training and development process.
The user for this module is the data scientist in our team supporting the customer and the consumer is the user on the customer side.
Objectives of the project:
- Conduct literature research to identify the state-of-the-art and possible technologies that can be used to build the prototype;
- Design an automatic feature extraction framework and process for time series data;
- Build an automatic feature extraction prototype that is specific for a predetermined set of problems that can extract useful features for a dataset.
- The module should be able to provide insights on why such features are selected. Insights should be automated and derived by statistical inference or explainable AI techniques
- Provide recommendations on how to connect the Automatic Feature Extraction module in a meta-learning for forecasting context
Project Deliverables
Automatic feature extraction module prototype including:
1. Sequence diagram of flow of interactions with the users and the other parts of the system
2. UX/UI prototype design (supported by our UX/UI team) as needed
3. Data Model for the specific module
4. Python processing logic in the form of background process or API
Tech Stack
o Python (prototype to be provided in either Flask, Streamlit or Dash)
o Tensorflow / Pytorch
You will be part of the Data Science team
About you
o BS / MS degree in Computer Science, Data Analytics or related disciplinary
o Familiarity with Python programming language
o Confident verbal and visual communicator
o A natural self-starter and problem solver
About us
We are scientists and engineers with a strong passion for cutting-edge technology and decision science. Our mission is to enable enterprises to reduce time to decision with state-of-the-art forecasting and optimisation. Our algorithms are based on automatic machine learning techniques and allow for the discovery of optimal forecasting models. We unlock sizable business and financial impact in terms of lower lost demand, decrease in inventory, lower working capital and financing costs.