Hiring

A/B Street is hiring a Rust engineer!

  • Timeline: Full-time for 3 months, starting ASAP. Latest possible start date is 1 June.
  • Compensation: ~£7k total
  • Must be UK-based with a UK bank account

The project

A/B Street aims to get the average citizen more involved with local transportation planning and accelerate plans to make cities more friendly to people biking, walking, and using public transit. This massive undertaking involves building a realistic model of any city's street network from open data, designing a user interface to easily edit streets and intersections, getting a traffic simulation to run with some degree of realism, using data visualization to explain the impacts of changes, and advocating for real changes using results from the software.

This position will support the sptreets project to explore scenarios of walking and cycling uptake and infrastructure in São Paulo, Brazil, and a project to support deployment of a strategic cycling uptake model. Depending on your skills and interests, there are a few possible places to focus for 3 months.

OpenStreetMap network simplification

Representing street networks geometrically from OSM data is incredibly hard. There are many limitations dealing with sidewalks, crossings, dual carriageways, and cyclepaths parallel to the road. These problems impact all parts of A/B Street -- for example, the traffic simulation tends to get stuck near complex junctions, and the low-traffic neighborhood tool often can't handle boundaries near overlapping road geometry. If you enjoy working with messy and incomplete spatial data, there's plenty of opportunity to make impactful progress here.

Further reading:

Public transit

A/B Street has yet to focus on public transit. Today, there's only preliminary and broken support for importing bus routes from GTFS data and simulating people using them. A vision of what proper public transit support in A/B Street should do:

  • Let people draw entirely new bus and light rail routes, then understand the impact on individual people and the aggregate community.
  • Explore how small changes affect bus performance -- like changing bus stop locations, adding a bus-only lane, or configuring a traffic signal to prioritize buses.
  • Gamify the process of planning for public transit, by gradually introducing editing tools and a budget, to teach the public about the trade-offs involved with planning.

In 3 months, you'd be expected to make progress on some tasks such as:

  • importing route and schedule data from GTFS
  • modifying pathfinding to decide which route a person should use
  • defining and visualizing metrics for performance of a bus route
  • and changing the map model and UI to allow bus stops and routes to be created and edited

Sharing proposals

After users modify a street network in A/B Street, they can save their changes as files locally or in their browser's storage. There's extremely basic support to anonymously upload the edits and share a URL with someone else. We'd like to have a proper way for people to share their proposals, with user accounts and versioning. Possibly we don't need to build, productionize, and operate our own service for this (and thus take on the burden of GDPR, handling abusive accounts, etc) and can instead leverage something existing.

Relatedly, we'd like to explore incorporating the web version of A/B Street into blog posts and story map-style articles, making it easy for campaign groups without much technical skill to embed interactive demonstrations.

Integration with other tools and deployment

An example of the benefits of integrating A/B Street with other tools is the ActDev project which consists of a web map embedded in a JavaScript powered dashboard front-end and a 'Simulation Layer' that can be clicked, taking the user into a dynamic simulation for each of the case study housing developments. Further scaling and integration challenges are presented by the projects that will fund this software engineering and development work

Your qualifications

  • A passion for open source software development enabling more evidence-based investment in transport solutions for a more sustainable future 🌱
  • Experience with Rust (or ability to learn Rust rapidly), an interest in transport datasets (e.g. OpenStreetMap, GTFS, origin-destination data). If you already know these, great. If not, you should feel comfortable learning enough to start contributing to the code-base in a few days.
  • Can work independently, but still communicate effectively with the rest of the team. We need to focus our efforts on other parts of the project, but we'll keep up with your work closely.
    • One way to demonstrate this: have you created your own project and maintained it for a while? Have you worked on other open source projects?
  • Are fine with (and happy to have) all of your work being Apache/AGPL licensed and designed/discussed transparently on Github and Slack/Discord.
  • Bonus if you have enough design sense to mock up UIs, but no worries if not.

Applying

If you think you'd be a good fit, email dabreegster@gmail.com and R.Lovelace@leeds.ac.uk with any relevant material -- open source projects, a resume, etc. We can schedule a video call.

In lieu of a traditional interview, we'd like to ask you to submit a PR to A/B Street to make progress on one of our starter bugs or any other fix/feature you'd like to see (there's plenty of things that need improving, and being able to find and fix them without much guidance is what we're looking for). Please focus on good communication in your PR, as it's something we'd expect you to keep up the entire time -- see this as an example, where the before/after of the change is clear, and any uncertain design decisions are brought up.