Link Search Menu Expand Document

The sections below are an attempt at summarising in an hopefully understandable and meaningful way what should be overall common traits of the backend group’s way of working, so that we can aim to be more intentional about how to represent our group’s work to potential candidates, whether in job descriptions or during job interviews and any follow-up questions.

Team collaboration

In most projects of small to medium complexity and lasting from a few weeks to a few months, backend development is typically done by a single engineer.

In more complex or longer projects, a second backend engineer may be part of the team throughout the project or for part of it. Occasionally we may also add freelancers depending on load and schedules.

Whether as sole backend engineers or with one or more team members, we work in very close contact with user researchers, designers, frontend engineers, scientist, data engineers and project managers, both through scheduled stand up meetings and sprint planning and retrospectives and by proactively reaching out to plan ahead, discuss functional design, unblock other project members’ work.

We may occasionally pair up remotely to work together on specific functional design tasks, to develop features, triage and fix bugs and to seek feedback and advice, although we don’t use pair programming as a daily way of working.

We take code reviews both as an opportunity to improve our projects and to learn and mentor.

We meet twice a week for two distinct types of team meetings. The first one is an organisational one, dedicated to updating each other and making team decisions broadly related to running a self-managed organisation. The other one is a technical meeting, where we share interesting bits of our daily work. We do also share frustrations, celebrate achievements and in general sustain the team’s growth, while keeping a light-hearted spirit. Once a month that meeting is held together with the frontend engineers team.

We try to make sure colleagues don’t feel stuck or concerned about daily work or longer term career development, by reaching out, listening, sharing opinions and experiences, or connecting with other team members who may be best suited to help.

We acknowledge that each colleague will have more experience with specific frameworks and technologies and less with others: we don’t do ego trips and we understand that seeking help or advice is a normal part of daily work and of organic team growth. Helping others helps one improve their own knowledge by forcing us to make knowledge understandable and meaningful.

Even before recent lockdown times, most of the backend team was fully remote: we expect team members to be able to work independently while keeping project colleagues up to date on progress and blockers and in general to work efficiently and independently wherever they may be working from.

Our daily work

Depending on project types, we may work on systems ranging from customized CMSes to APIs (mostly REST, less frequently GraphQL).

Our APIs typically support SPA frontend apps based on React and NextJS, and a range of data visualization and mapping libraries.

We use both monolithic and microservices architectures - again, depending on projects.

Our core frameworks are Ruby on Rails, Express, NestJS. Most of us specialize in one framework but we try to develop a basic familiarity with the rest so that we can jump in to help when needed.

Most of us are also conversant with Python: the Science team uses it as one of the core languages, and we help maintain microservices that typically use the Flask framework.

Especially as some of our projects are expected to evolve through the years, we strive to make our code understandable to our future selves or to colleagues who may work on them in the future. This includes applying architectural patterns, writing essential and accurate documentation in English, and aiming for clean code in general.

Testing discipline varies across projects, but we aim to have coverage at least for core features and code paths.

We typically design, set up and maintain backend and data infrastructure of our projects, ranging from single VPSes to Kubernetes clusters, across staging and production environments.

We test and deploy most of our code via CI/CD pipelines. We may run key tests via git hooks.

We may use Docker and Docker Compose for development environments.

We run most of our projects on a range of cloud services typically either on the AWS or GCP platforms.

We typically take care of database design, operations and performance. We work with relational databases (mostly PostgreSQL), NoSQL (mostly MongoDB), and other data stores such as Redis.

Most of our projects are relatively data-intensive, so we expect all team members to have or develop a strong command of advanced data management techniques.

In some projects our work may intersect with that of scientists and data engineering. We design and operate ETL pipelines and occasionally help with ML workflows. We expect this to become an increasingly central part of our work in larger projects.

We expect team members to take ownership of projects beyond the backend area, by proactively working with project teams and clients to make sure that the data, API and infrastructure parts of projects effectively support each project’s business requirements.

By virtue of our ownership of the foundational layers of most projects, we expect team members to provide clear, thoughtful and accurate advice to the other functional areas about alternatives for the implementation of major features, with special attention to performance, developer ergonomics, operational costs, ongoing maintenance and possible future evolution.

Occasionally we collaborate with the Business Development team on project proposals. We may write the backend-specific parts of proposals, estimate scope, validate budgets.

Our values

We value thorough and broad expertise beyond our core work domains, intellectual curiosity, eagerness to learn, ability to mentor others.

We value the willingness to take measured risks, to seek advice on risk, and to help others manage risk effectively and healthily.

We value the ability to communicate clearly and fluently in English both with team members and with clients and non-technical people.

We value honesty, emotional intelligence, passion for craftmanship and eagerness to improve the quality and impact of our work.