Product Data Scientist
Job DescriptionJob Description
Job Description
We are Baton Corporation, a team of elite builders, moving fast and redefining whatʼs possible. Weʼre a high-growth software development company and proud contributors to pump.fun, the largest crypto social network globally. We build for the future.
Baton Corp& is a pioneering platform that enables anyone to create a coin in under 1 minute for free. Launched in January, the platform has quickly become the home of memecoin creation and trading on Solana, and by extension, the entire crypto space. The platform has generated over $550 million in revenue since March 2024 - attracting significant attention in the crypto community.
We are in search of a high-performing Data Scientist who is passionate about extracting meaningful insights from complex data sets. This individual will play an integral role in helping us understand our growing data, which will directly influence our strategic decisions. The ideal candidate is nifty, detail-oriented, and has a keen interest in exploring and unraveling complex data trends. Familiarity with the world of cryptocurrencies is a plus, but not a requirement.
In general, we want an experienced, flexible, and driven data scientist who has a broad range of knowledge in the field rather than a specialist who has focused on a specific area. They need to be autonomous and self-motivated, who can work with other teams to measure the impact of changes, as well as identifying problems, then developing and implementing solutions.
Requirements:
- Proficient with Python/R and SQL/Bigquery.
- Experience designing, running, and analysing A/B tests (experiments) on a tech consumer product.
- 2-3 years Minimum of experience at a startup, scale-up, or FAANG/equivalent.
- Demonstrated experience of autonomy and agency in overseeing entire projects from identifying problems, testing solutions to communicating results.
- Can work full time in NYC 5 days a week.
:
- Experience with crypto and blockchain data.
- Strong data visualisation skills (esp. Graphana)
- Experience with events tracking data (esp. FullStory)
- Strong statistical or causal inference skills.
- Experience building data pipelines (esp. Dataform)
- Experience with Machine Learning methods.
Additional notes for Product Data Scientist:
Autonomous and Self-Motivated: here we want someone who is going to be pro-active (identify and find work) rather than reactive (be assigned work). Either reaching out to other teams to help them with measurement of changes, or who can spot problems/opportunities and drive forward changes needed.
Specialist vs Generalist: ideally here we want someone who isn’t too excessively focused on a specific aspect of Data Science (e.g. only ML experience, only experience with non-experimental methods)
Consumer Product: this means basically any work that was focused on an app or website, rather than more general data science (finance, health, education, economics, etc.). When the data scientist was involved in measuring changes to the app/website.
Overseeing Projects: what we mean here is that the person has experience managing projects directly. From the ideation phase, to prioritisation, stakeholder management, design/analysis, and rollout.
Data Visualisation: this would involve building and automating dashboards (experience with Graphana is a plus here), but also being able to visualise data analysis in a clear, concise, digestible format.
Racking Data: this type of data captures how a user moves throughout (and interacts with) our website: the consumer funnel. What they clicked, what sessions they had, did they place an order, did they experience an error, etc.. Our tracking data comes from FullStory so experience with it is a plus.
Statistical or Causal Inference: these capture non-experimental methods of data analysis. Essentially, can this person give us learnings and insights from non-experimental (observational) data. This work is often the type of thing that starts experimental work: you see something interesting, dig into it and find the relationships, build a hypothesis of what is going on and then run an experiment to test that. It would also include econometric modelling methodologies (using correlations to model the impact of a change).