About the role

Maincode is training the next version of Matilda, the first large language model built and trained from scratch in Australia. We are now scaling the model and deploying it as a live production system.

To do that, we build AI systems from first principles. We design the architectures, run the infrastructure, shape the training process, and operate the systems that serve our models.

Matilda is a production system, not a research prototype, built to be deployed and served for open public access. If the training stack is how Matilda learns, the product is how people experience it. This role sits directly in that delivery layer.

You will build the front end systems that make Matilda usable, fast, and trustworthy for real users. You will work on the interfaces that turn a model into a product: streaming responses, conversation state, safety UX, performance, reliability, and the tooling that helps the team iterate quickly.

What you would actually do

You will build and maintain the front end systems that power Matilda’s public product and internal tools.This includes:

• Building the core web interface for interacting with Matilda (chat, sessions, history, settings)

• Implementing fast, reliable streaming UX for model outputs (real-time tokens, partial results, cancellation, retries)

• Designing and building UI workflows that support safe and responsible use (reporting, refusals, user feedback, content handling)

• Working closely with backend and infra engineers to integrate APIs, auth, rate limits, and observability

• Improving performance across the app (latency, bundle size, rendering, perceived responsiveness)

• Building internal dashboards and tools that help the team operate the system (usage, quality signals, feedback triage)

• Debugging issues across browsers, devices, networks, and production environments

• Raising the bar on quality through testing, monitoring, and careful rollout practices

You will spend time in code, UI state, performance traces, and production metrics.

The kind of person who does well here

We are looking for engineers early in their careers who want to build the product surface of a frontier model, and who care about quality and reliability in real production environments. You may have one or two years of experience building production software. What matters most is curiosity, taste, and the willingness to learn how these systems behave with real users.

People who tend to do well here:

• Care about performance, UX quality, and edge cases that show up in production

• Enjoy building clean, maintainable front end systems, not just one-off screens

• Have strong debugging instincts and do not panic when something breaks

• Think clearly about product tradeoffs, reliability, and failure modes

• Pay attention to metrics and user feedback, not just what looks good locally

• Want to work close to the core system, not just build a marketing site

You do not need prior experience working on AI products. What matters is intellectual curiosity, persistence, and the ability to learn quickly.

How you would work

You will write production code that sits in the product layer of the Matilda stack. You should be comfortable:

• Working in TypeScript and modern front end tooling

• Building production web applications (component architecture, state management, API integration)

• Implementing real-time or streaming experiences (for example, via SSE or WebSockets)

• Debugging complex UI state and issues that only occur under real conditions

• Collaborating closely with user research, backend, infrastructure, and research teams

Much of the work sits between product and the underlying model. The model evolves quickly, but the user experience must remain stable. You will be working on the front end systems that ship Matilda to the public and help operate it day to day.

Why Maincode

Maincode builds AI systems end to end. We prepare the data, design the training process, run the infrastructure, and operate the models ourselves. You will work with a small team that:

• Builds the full AI stack rather than outsourcing it

• Treats product quality and reliability as part of the intelligence system itself

• Values engineers who want to understand how things actually work

• Is building long-term capability in training, deploying, and operating large models

If you want to build the product surface of a model trained from scratch, and ship it to real users, you will be close to the core work here.

Note

This is a full time role based in Melbourne, working closely with our in person engineering and research team. At this time we are not able to offer visa sponsorship, so applicants must have existing and unrestricted work rights in Australia.