Free
$0
Read all briefs and paths. Tool stages and written work are locked.
Keep browsingProjects/Wayfriend
From invites to meetups
Wayfriend wants to know whether friends who plan to meet actually spend time together. Clean the raw events, model meetups, measure success and plan the next day of work.
One brief, four toolsYour tables pass from Databricks to dbt to SQL, then to Dagster. You finish with a written plan.
The brief
Wayfriend's product lead, Rosa Imani, has a question for the data team: when two friends accept a plan to meet, do they actually end up together? She wants a measure the team can trust and a plan for what to look at next.
You have two weeks of location pings, invites and users, logged in UTC. An accepted invite counts as a meetup when both friends pinged within 50 meters of each other in at least 10 separate minutes during the two hours after the planned start.
The raw data is messy in the ways app data usually is. Some pings were resent with the same id, some have no coordinates, and invite statuses were typed in by three app versions. Location data can show that people were together; it cannot show why.
- Data window: October 1 to 14, 2026, in UTC.
- You deliver clean tables, a tested meetup model, three success measures and a plan for tomorrow.
- Checks cover duplicates, missing coordinates, invites with no pings and the meetup rule.
The project path
Open any available stage- 1Databricks40 min
Clean the raw events
Load the pings and invites with PySpark. Remove resent pings, drop pings without coordinates and tidy the invite status.
- Reads
- raw_pings · raw_invites
- Produces
- clean_pings · clean_invites
dbt reads these two tables.
Open stageOne Course in Databricks or All Access - 2dbt50 min
Build the meetup models
Model every accepted invite and whether it became a meetup. Test the invite key so a duplicate can never reach the measures.
- Reads
- clean_pings · clean_invites
- Produces
- fct_meetups
SQL reads this model.
Open stageOne Course in dbt or All Access - 3SQL45 min
Measure meetup success
Calculate the meetup rate and the median time to meet, city by city.
- Reads
- fct_meetups · users
- Produces
- meetup_metrics
Dagster schedules these results.
Open stageOne Course in SQL or All Access - 4Dagster50 min
Keep the pipeline running
Partition the meetups by day, schedule the daily run for 03:00 UTC, and remove location pings after 30 days.
- Reads
- fct_meetups · meetup_metrics
- Produces
- daily_meetups
Your plan uses the daily run.
Coming soonWorkspace in progress - 5Plan25 min
Write tomorrow's plan
Explain the result, name a limitation, choose the next investigation and say how you will judge success.
- Reads
- meetup_metrics · daily_meetups
- Produces
- next_day_plan.md
Optional: present your plan in Interview Hub.
Open stageAll Access
Each stage includes reference inputs. Starting in the middle or using them doesn't complete the earlier stages.
Project access
One Course
$17.99/month · USD
Practice the stages in your chosen tool. Reference tables let you start at that stage.
Choose a courseAll Access
$24.99/month · USD
Every stage in every format, plus optional spoken readouts using plates.