About the job
About the Role
Maincode is on a mission to develop Matilda, a cutting-edge large language model created entirely in Australia. With our new compute cluster now operational, we are scaling the next iteration and preparing for public deployment.
This position is integral to our production system that supports Matilda. You will be tasked with creating and maintaining back-end services that ensure the model's practical usability, including APIs, infrastructure, and systems that transform a trained model into a dependable public resource.
Our approach to AI development is comprehensive; we architect the systems, manage the infrastructure, train the models, and operate the solutions ourselves. Matilda is not merely a research prototype; it represents a robust production system trained at scale and made available to the public.
Maincode boasts one of Australia's largest private AI compute environments, dedicated to the training and operation of our proprietary models. You will directly engage with the systems that deploy and serve a model built from the ground up.
Your Responsibilities
Your role will involve building and sustaining the services that bridge the model and external interactions.
This encompasses tasks such as:
- Developing and maintaining services for model inference and user requests
- Designing systems for managing requests, sessions, and streaming responses
- Implementing reliability features including rate limiting, retries, and graceful failure mechanisms
- Establishing authentication and access controls for public use
- Creating systems for logging, telemetry, and evaluation signals
- Enhancing latency, throughput, and reliability in model serving
- Incorporating new model checkpoints into the production environment
- Collaborating closely with training and infrastructure engineers to deploy and maintain the model
Most of your work will occur within production systems, focusing on logs, traces, performance profiles, and deployment pipelines. Our priority is to ensure a system that remains operational, efficient, and predictable under load.
Who Thrives in This Role?
We seek engineers at the beginning of their careers who are eager to learn how to build and operate production AI systems effectively.
You may have one or two years of experience in developing production software. What we value most are curiosity, reliability, and a willingness to understand large-scale system operations.
