This is an exciting opportunity to be part of a brilliant team in a fast-paced, collaborative environment. You will have the chance to influence and shape our Product, Technology and Data Science strategy while working with billions of data points to solve real-world, measurable problems!
CitySwift is a Cloud-native, specialist data engine for modern bus networks. We optimise urban bus networks using Big Data and Analytics. Ultimately, we improve the reliability of services while simultaneously reducing Operator costs, resulting in a win-win for both passengers and operators!
Our Company Values
- Be Open & Honest
- Take Ownership & Finish it!
- Think like a Customer
- Alright is not OK!
- Be up for the challenge.
What You’ll Do:
- Build statistical and machine learning models to understand and predict various aspects of public transportation, including demand, run times and usage patterns to name a few.
- Conduct rigorous testing and evaluation of existing models, suggesting improvements and optimisations as well as exploring alternative approaches.
- Conducting R&D and exploratory analysis into new avenues of interest, with an aim to better understand the problem to be solved and gain deeper understanding of the nuances of data driven public transport optimisation.
What you’ll bring:
- Proven experience of data science and machine learning applications in a variety of areas and solving different problems using a variety of techniques.
- Solid grounding in statistics and with proven ability to relate statistical analysis and metrics to real world problems during product development.
- Strong Python 3.x knowledge with continuous use of pandas, sci-kit learn, TensorFlow and stats models or any other model building packages.
- Strong SQL and data manipulation skills.
- Experience building neural networks using TensorFlow bespokely and using the estimator API.
- Experience using cloud based platforms for data manipulation but also cloud based model training and hosted real time and batch predictions.
It would be great if you have:
- Experience with large scale data pipeline for leveraging external sources eg. Apache beam
- Experience of model specific transformation and feature engineering pipelines such as sci-kit-learn pipeline or TFX.
- Experience with both TF1.x and 2.x with detailed knowledge benefits of both.
- Experience of model explain-ability techniques.
- Experience developing routing based predictions or area specific demand