Senior Data Scientist II, Machine Learning

Data Scientists at SpotHero create data products fueled by machine learning under the hood. Over the course of your career at SpotHero, you will act in an end-to-end capacity to own and mature a portfolio of data science solutions. You will work across teams to understand needs, create data science solutions, advise stakeholders, and design advanced experiments to validate ideas and models.

As a Sr. Data Scientist II, you will be expected to carve out a portfolio of investigations and insights based on a broad need of ours. This will involve pairing with product managers and/or product designers to conduct discoveries to understand more about the domain stakeholder needs. You will then author problem statements with actionable hypotheses based on your discovery. After this, you will prototype solutions, work with stakeholders to test these prototypes, and improve / iterate on them as needed. If the acceptance criteria are met, you will then mature the solution into production working with our Machine Learning Engineers. Some examples of our broad themes include:

1. Pricing: Can we measure price elasticity by analyzing our historical data? Is it possible to measure price elasticity at a given parking facility? What factors influence price elasticity?

2. Personalization: How do drivers use our app and can we glean any insights about their buying preferences or parking intent? What opportunities are there to personalize the search experience? To what extent do price and distance alone explain why drivers choose one facility over another?

3. Business Operations: What is the cost of acquisition and lifetime value of our customers? How can we optimize marketing spend to reduce the cost of acquisition? What markets should we expand in?

In addition to focusing on one of these themes, you will be expected help with ad hoc requests that arise from stakeholders (such as advancing our dynamic pricing solution), provide constructive feedback on squad deliverables, and suggest how we can improve as a team. Everyone on the Data Science squad also invests 20% of our time proposing and investigating our own questions.

Overall, this role is a great opportunity if you are an experienced data scientist who is looking to put into practice the skills and experiences you’ve learned to carve out a portfolio of data science products. This is also an opportunity that will appeal to those who want to learn from a diverse team of subject matter experts spanning various engineering, business, and product domains.

Who we are: 

At SpotHero, we work as a team to empower people to get everywhere, easier! We’re rapidly growing with the mission of bringing the parking industry into the future through technology. Drivers across the nation use the SpotHero mobile app or website to reserve convenient, affordable parking in advance, on-the-go or through their connected cars, and parking companies rely on us to help them reach new customers while optimizing their business. We connect the dots with cutting-edge technology, delivering value to both sides of this exciting, evolving marketplace..

What will you do:

Modelling and Engineering (45%)

• Own the implementation, deployment, and maintenance of data science models and design experiments to validate their performance
• Apply and suggest improvements to our technologies and best practices so that we can validate our ideas faster and create scalable solutions
• Give constructive feedback on squad deliverables, workflows, and our tech stack to improve the quality of our work
• Write concise, modular code that handles data cleaning, data aggregation, data exploration and statistical or machine learning modeling
• Keep a pulse on and surface relevant industry news, best practices, models, and technologies that could be used to solve our problems

Business and Storytelling (20%)

• Write problem statements that breakdown broad, ambiguous, complex needs into actionable questions and hypotheses
• Present deliverables to interdisciplinary audiences with an appropriate balance of technical and business acumen
• Proactively share your recommendations around next steps and effort estimates
• Author documentation that is clear and accessible to its intended audience
• You help create a data science community by sharing your thoughts on data science opportunities, speaking at sprint demos, and representing SpotHero at external conferences

Data (20%)

• Prepare and structure custom data sets spanning data from across SpotHero for bespoke analysis and authoring proof of concepts
• Incorporate and expand on existing data models
• Analyze large amounts of data to uncover trends, patterns, and insights
• Author new KPI’s that help us size opportunities and benchmark current performance

Statistics / Experimentation (15%)

• Design and take feedback on creating experiments to test and evaluate the performance of models
• Review and give feedback on experiment proposals and analysis

The following experience is relevant to us:

• Must have a portfolio of data science use cases you have created and/or contributed to that shows your breadth and depth as a Data Scientist. Ideally, this portfolio is a balance between early proof of concepts through to late-stage production solutions
• Exposure to facilitating discoveries to understand more about a need and experience contributing to writing problem statements, hypotheses, and/or business cases around data science use cases
• Adept at communicating with a diverse group of stakeholders on a variety of business and technical subjects
• Experience contributing to software (pipelines, services, and client libraries) that is run in Docker containers, automatically tested on a continuous integration (CI) system, and versioned in git
• Working knowledge of Data Science analysis and modeling packages in Python including Pandas, Scikit-Learn, NumPy
• Good understanding of a SQL dialect to query, read, and write relational data
• Exposure to MLOps platforms and technologies such as Databricks, Sagemaker, Domino, ClearML, Pachyderm, etc. 

Nice to have:

• Demonstrated experiences with designing and conducting experiments to evaluate machine learning models
• Experience with multiple data platforms and tools (eg. S3, Redshift, Airflow)
• Experience using distributed systems (eg. Apache Spark, Presto, Dask)
• Experience writing shell scripts, Makefiles, or other configurations to glue together these components
• Familiarity with BI solutions like Looker (preferred), Tableau, Sisense, etc. 

Technology we use:

• Our Android Stack is: Kotlin and XML (standard for Android apps) using MVI architecture (still working on refactoring old views), our database layer is built in Realm. Bitrise for CI/CD. We also make heavy use of Dagger, RxJava, Espresso (testing). Network stack uses Retrofit.
  
• Our iOS Stack is: Swift using MVC architecture, CoreData for Local Storage, XCUI for UI Testing, XCTest for Unit testing, SPM for Package Management, Fastlane for app automation and build scripts, Bitrise for CI/CD, and Sentry for crash reporting.
  
• Our Back End Stack is: Monolith using Django/Python/PostgreSQL. We are moving our Monolith to a Modular Monolith, using Domain Driven Design. When relevant we extract specific domains to Services currently using Java, Kotlin and Go. We also use Docker, deploy our apps via Kubernetes. We use Kafka for asynchronous-, and gRPC for synchronous service-to-service communication. Our Integrations are on a .Net CORE, moving to Kotlin.
  
• Our Front End Stack is: Our Front End stack is React/Redux, Sass, Jest/React Testing Library/Cypress, and Webpack. We maintain a private npm repository with shareable UI components, utility functions, Babel/ESLint/Prettier configurations, and custom tasks
  
• Our Data Stack is: Our Monolith Database is Postgres and Redis for caching. We also use Redshift as our data warehouse and S3 as our data lake. The data lake is queried using Presto. We use Airflow and Spark for ETL, as well as do some stream processing (Kafka Streams and Spark at the moment). Our Model pipeline uses scikit-learn, pandas. Our analysts utilize Looker as our Business Intelligence tool. And we use Quicksight for Dashboard on our external Data Products.
  
• Our Dev Tools Stack is: AWS+Kubernetes for hosting. Terraform + Helm Charts for IaaS/Deployment. ConcourseCI for CI/CD. Prometheus/Alertmanager/VictorOps for team alerting. We’re starting to work on multi-region available services.

What we are offering:

• Career game changer – A truly unique experience to work for a fast-growing startup in a role with unlimited potential for growth.
• Excellent benefits – 
• In the US we cover up to 95% of Medical Premiums, 50% of Dental & Vision Premiums, company sponsored Life Insurance, and generous parental leave. 
• In Canada we offer Medical (prescription drug and paramedical coverage), Dental, Vision, Life Insurance, STD and LTD. 
• Flexible PTO policy and great work/life balance – We value and support each individual team member.
• Annual parking stipend – we help people park!
• The opportunity to collaborate with fun, innovative, and passionate people in a casual, yet highly productive atmosphere.
• A workplace recognized as the Best Consumer Web Company by Built in Chicago, Top Company Culture by Entrepreneur, a Top Workplace by Chicago Tribune, and one of Chicago’s Best Places to Work for Women Under 35 by Crain’s Chicago Business. 

Steps to apply: Please include any GitHub account, LinkedIn profile, and any project that you’re particularly proud of. We love seeing work that others loved working on.

At SpotHero, we Respect Fellow Drivers by providing an inclusive interview experience for everyone, including people with disabilities. We are happy to provide reasonable accommodations to candidates in need of individualized support during the hiring process. Please let our team know of your need when you apply or as you begin interviewing with our team.

SpotHero is an equal opportunity employer. We know that a diverse workforce is the strongest workforce, and are committed to building and supporting an inclusive environment for all.