Senior Ai-Native Product Engineer

Core stack: Elixir, Phoenix and LiveView

If you’re a senior engineer with strong fundamentals looking to retool for AI-native development - or already an AI-native developer and looking for an environment designed around it - read on.

Cheerful Monster is a London-based development studio building software with a smile. We deliver world-class AI and software for clients; make developer experience and career building a first class activity; run experiments to redesign the nature of software delivery in the AI enabled world.

We combine product thinking, technical strategy and software delivery in one small, senior team. Rather than passing work down a chain of specialists, our engineers work directly with clients and own outcomes from discovery to production.

We’re hiring a senior engineer to deliver client work now and help establish the practices future project pods will learn from. This is a full-time role, working from our Canary Wharf office in London, some hybrid flexibility is available.

The role

You’ll work as a product engineer in a small, mission-driven pod, directly with clients, taking problems from early ambiguity through design, implementation and deployment to production outcomes.

Our first projects will use Elixir, Phoenix and LiveView . You should have used this stack on at least one production system, then that is great. If not then you must be fluent in several other high level languages, so that you can easily assess the output of coding agents.

Coding agents will be part of the normal development environment: exploring codebases, developing plans, implementing features, generating tests, reviewing changes and investigating production faults. Ideally you have experience using skills, tools, plugins etc, if not then you are keen to learn and will lean in to up skilling in this area.

You remain accountable for the agent's output and whether it is fit to deploy. AI makes writing code cheap, but it cannot decide what should be built, understand the client’s situation, make sound architectural trade-offs or establish that plausible-looking output is actually correct.

What you will do

You’ll turn ambiguous client needs into clear product and technical direction, separating the outcome a client needs from the first solution someone happens to suggest.

You’ll design and build production systems with Elixir, Phoenix and LiveView, working across domain modelling, data, interfaces, integrations, background processing, testing, deployment and observability as each project requires.

You’ll decide what to do yourself, what to delegate to agents and what evidence is needed before the result can be trusted. Sometimes that will mean running investigations in parallel; sometimes discarding generated work; and sometimes deciding that AI is not the right tool for the job.

You’ll also review and develop other engineers. As one of the first senior engineers working within Bonfire, you’ll help establish how we capture context, structure work, verify generated changes, manage cognitive load and demonstrate increasing autonomy. You’ll contribute to our shared engineering practice and the development harnesses that support it.

And you’ll have fun. The things that made programming enjoyable haven’t disappeared; they’re wearing new clothes. Part of what we’re building is a way to find the craft and joy in software development again.

Requirements

You may already use coding agents as a routine part of production delivery. You can explain how you frame work, supply context, delegate meaningful implementation, review what comes back and catch output that is convincing but wrong.

Or you may be a strong senior Elixir engineer whose organisation has not yet given you the opportunity to work this way. You have begun experimenting seriously with AI-assisted development and want to build a more disciplined approach to specification, delegation, verification and agent orchestration.

We regard both as first-class routes into the role. In either case, we’re looking for strong engineering judgement, production experience and a willingness to make your working process visible and improve it.

You should have substantial experience building and operating software with Elixir and Phoenix, with a strong grasp of OTP, supervision, concurrency, Ecto and relational data modelling. You should be able to spot generated Elixir that compiles but is unidiomatic, poorly structured or based on a mistaken model of the runtime.

Product judgement matters as much as stack depth. You should be comfortable working directly with clients, shaping ambiguous work, explaining trade-offs and taking responsibility for what happens after deployment.

Clear writing matters too. Plans, specifications, constraints, acceptance criteria and review comments are increasingly part of the engineering interface between clients, colleagues and agents.

Technology and useful experience

The initial environment centres on Elixir, Phoenix, LiveView, OTP, Ecto, Ash and PostgreSQL , alongside the usual testing, Git, CI/CD, cloud deployment and observability tooling.

Experience in other languages is welcome but not required. Python is particularly useful , whether you’ve used it for services, data, automation, scientific computing or AI integrations. Experience in another ecosystem is also valuable where it demonstrates real depth and an ability to move intelligently between technologies.

Experience with Oban, Broadway, distributed Elixir, data engineering, LLM integrations, regulated domains, consultancy, early-stage products or coaching other engineers would all be useful. We don’t expect one person to bring all of them.

• Company pension
• 25 days of paid annual leave (pro-rata)
• A fun and flexible hybrid working environment
• A modern, dog-friendly office located in Canary Wharf

Senior AI-Native Product Engineer • London, England, GB

#J-18808-Ljbffr