Internships

incident.io has an internship programme for student engineers who want to gain work experience over a 6 month placement.

You can apply here or read about the role below.

Caution

Applications are now closed. Will re-open again for Spring 2025 placements around October 2025.

What is the role?

This placement is for a Product Engineer role: watch this video to see how it differs from other roles.

We're looking for candidates that:

• Want to work in a fast-paced engineering team in one of London's fastest growing start-ups.
• Love the idea of working closely with customers to build a product used daily to respond to critical incidents.
• Are excited by the opportunity to build software used by some of the most recognisable names in tech: Etsy, Skyscanner, Vercel, Linear and more that we can't share yet!

Do not apply if:

• You're looking for an 'intern' project: this placement will have you doing real work and we'll make little distinction between an intern and a fulltime engineering hire.
• You don't enjoy learning new technologies by 'doing': we expect engineers to get involved building and learn as they go (most have learned Go while on the job, as an example).
• You prefer to work solo: we're highly collaborative and all your work will happen as part of a team.

Application process

Please apply via our Product Engineer Intern job posting. When applying, include a CV detailing education and past experience and a cover letter that explains why incident.io is particularly exciting to you.

We care more about the cover letter than the CV, so think carefully about why this role is something you would do well in! Make sure to read What is this role? as we're looking for reasons you would be a good match against that description.

Pro-tip: you can get a sense of engineering at incident.io by reading our engineering blog.

The full candidate process will be:

1. Apply with cover letter.
2. If accepted, we'll ask you to provide either:
   • Existing product you're proud of.
   • Solution to our take home test.
3. If successful, we'll invite you for an on-site interview on the 5th December 2023.

Please apply via the job link and we'll share everything you need to know, but for more detail on step (2) read below:

1. Existing project you're proud of

Have you built something you're proud of, or think showcases skills that would be valuable for this role?

If so, we invite you to submit this project in leiu of a take-home challenge.

An ideal project would:

• Be a software project that solves a real problem.
• Demonstrate your ability to build high-quality software.
• Provide enough context (often through a README) that would allow anyone grading your code to get setup and run it.

We'll only accept submissions where you are the primary contributor to the project: group projects are great but become extremely difficult to evaluate fairly!

2. Complete our take-home challenge

Instead, you could try solving the following code challenge.

Intro

Imagine you're working at incident.io and we're building a product that can page (call them, send an SMS, send a notification to a mobile app, etc) engineers when their services are involved in an incident.

When configuring an on-call system, you don't want to say "whenever service X goes down, page Y person" as that person probably has a social life and won't appreciate receiving all the pages, all the time.

Instead, you want to build schedules: a set of people who take it in turns to provide cover for a service by rotating through on-call shifts.

In JSON form, the configuration that describes how a schedule behaves might look like this:

// This is a schedule.
{
  "users": [
    "alice",
    "bob",
    "charlie"
  ],

  // 5pm, Friday 17th November 2023
  "handover_start_at": "2023-11-17T17:00:00Z",
  "handover_interval_days": 7
}

In that example, our schedule will rotate evenly between those users with the first shift starting at 5pm Friday 17th, with shift changes happening every 7 days.

That means:

• Alice takes the shift for 1 week, starting at 5pm, Friday 17th November
• Then Bob is on-call for 1 week from 5pm, Friday 24th November
• Then Charlie, then...
• Back to Alice again.

Visually, this might look like this:

Schedule systems often support 'overrides' where you can add temporary shift modifications to a schedule, such as if someone wants to go walk their dog or go to the cinema.

An override specifies the person that will take the shift and the time period it covers. An example of Charlie covering 5pm-10pm on Monday 20th November would look like this:

// This is an override.
{
  // Charlie will cover this shift
  "user": "charlie",
  // 5pm, Monday 20th November 2023
  "start_at": "2023-11-20T17:00:00Z",
  // 10pm, Monday 20th November 2023
  "end_at": "2023-11-20T22:00:00Z"
}

Task

We would like you to build – in any language you choose, but ideally one you are very comfortable in – a script called ./render-schedule that implements a scheduling algorithm.

It should be run like so:

$ ./render-schedule \
    --schedule=schedule.json \
    --overrides=overrides.json \
    --from='2023-11-17T17:00:00Z' \
    --until='2023-12-01T17:00:00Z'
[
  {
    "user": "alice",
    "start_at": "2023-11-20T17:00:00Z",
    "end_at": "2023-11-20T17:00:00Z"
  },
  {
    "user": "bob",
    "start_at": "2023-11-20T17:00:00Z",
    "end_at": "2023-12-01T17:00:00Z"
  }
]

Where:

• --schedule JSON file containing a definition of a schedule (see above example)
• --overrides JSON file containing an array of overrides (see above example)
• --from the time from which to start listing entries
• --until the time until which to listing entries

The script should output a JSON array of final schedule as a list of entries. This should take into account the projected entries (based on the handover information in the schedule) alongside the provided overrides.

Your schedule should also be truncated based on the from/until parameters provided. For example, if an entry was from 1pm November 17 -> 1pm November 19th, but from was 2pm November 18th, the entry should be returned as 2pm November 18th -> 1pm November 19th (ignoring the part of the entry that is outside the provided range).

Entries should be truncated to match the from/until parameters.

Code submission

Once we've reviewed your CV, we'll send you instructions on where to send your solution. We'll expect a ZIP file containing the code, plus a README.md that includes:

• Instructions on running your code, including installation of dependencies.
• Comments on code structure/how you approached the problem you think may be useful.
• Having now built the scheduler, any thoughts you have on the product features you might build on top of this (if you think how we'd use this at incident.io, for example).