Role overview
- Work with the product team, end users and engineers to identify and understand the problems we need to solve
- Work with other experts to identify feasible solutions to those problems that meet business needs
Develop production ready models in areas such as
- Optimising programmatic auction bids
- Understanding the content where adverts are being served
- Moderating content to ensure it is safe for kids and teens
- Categorising content so we analyse it quantitatively and infer our audiences
- Predicting what categories will improve advertising performance or brand lift
- Identifying audience clusters from survey and performance data
Test and validate the models and monitor and maintain them once live
What you'll work on
SuperAwesome helps brands connect with youth audiences with compliant, privacy preserving and age-appropriate digital campaigns.
We are looking for a Data Scientist to come and help us deliver original solutions to new problems by understanding the content youth audiences are engaging with and the brands that are reaching out to them. We combine this with knowledge from our behavioural insights team to predict and plan campaigns with impact.
Through ensuring best practice, privacy and moderated safe content we help fund and power a better internet for the next generation.
If you like collaborating on difficult problems with noisy, non linear data and you tick some of the boxes below, we would love to hear from you about what you are great at and what you are passionate about.
What we're looking for
- Strong pure mathematics
- Bayesian modeling
- Causal inference
- Post graduate degree
- Interests in behaviour, psychology or advertising
- Experience in adtech, martech, marketing or advertising
- Experience with survey data or brand lift studies
- Experience building models for internal or business facing products
- Able to communicate what you are doing to a wide audience
- Full ML model lifecycle experience
- Programmatic use of LLMs
- LLM fine tuning
- Working with big data tools eg pySpark or similar
- Github or similar version control
- Jira or similar workflow management
- Using Cloud platform ideally AWS
- Clean modular code with tests
- APIs eg FastAPI
- MLOps experience