Senior Software Engineer, Machine Learning

Mary Technology; Sydney; On-site; AUD 120,000 - 150,000; Full time

Benefits offered by this job

Equity in the company

Career growth opportunities

Direct impact on product development

Job summary

A leading legal tech company in New South Wales is seeking a Senior Machine Learning Software Engineer. In this role, you will focus on fine-tuning and deploying transformer models, building scalable inference systems, and leading experimental initiatives to drive product innovation. A strong background in ML and transformer pipelines is essential for this position. Join us to transform how the legal industry operates.

Qualifications

• Experience training and iterating on production transformer models and LLM pipelines.
• Hands-on expertise with PyTorch, high-level ML frameworks, and inference engines.
• Self-motivated learner who stays current with ML research.

Responsibilities

• Fine-tune and iterate on transformer-based models.
• Build scalable, low-latency inference systems.
• Design synthetic data pipelines and simulation environments.

Skills

Production transformer models and LLM pipelines

PyTorch

High-level ML frameworks

Inference engines

Current ML research

Job description

About Us

At Mary Technology, we’re tackling one of the legal industry’s biggest challenges: Fact Chaos — the overwhelming disorder that arises from managing mountains of unstructured information.We’re building intuitive, scalable tools that help lawyers spend less time untangling facts and more time on high-value thinking. From visualising case timelines to extracting key insights from documents, we’re reimagining what legal work looks like.We’re backed by $2.7M in funding, growing over 30% MoM, and working with some of the largest law firms in Australia and 50+ mid-tier firms.If you're excited by AI, legal tech, and solving real-world problems at scale—we want to hear from you.

The Role

We’re looking for a Senior Machine Learning Software Engineer to drive forward both research and implementation. You’ll be working on cutting-edge models and infrastructure that transform how legal professionals interact with data.

You’ll fine-tune and deploy production-grade transformer models, build simulation environments for synthetic data, and design scalable inference systems — all while working closely with a high-performing, curious team.

Your Impact

• Fine-tune and iterate on transformer-based models — LLMs, embedding models, cross-encoders, and other components of modern LLM pipelines.

• Build scalable, low-latency inference systems with monitoring, optimisation, and dynamic scaling.

• Design synthetic data pipelines and Gym-style simulation environments to support model evaluation.

• Lead experimental ML initiatives that drive product innovation and long-term value.

• Help shape the foundational systems for Legal AGI.

What We’re Looking For

• Experience training and iterating on production transformer models and LLM pipelines.

• Hands-on expertise with PyTorch, high-level ML frameworks, and inference engines (e.g. vLLM, SGLang, TensorRT).

• Profound grasp of the forefront of ML and LLMs, from weight quantisation to the nuances between MLA and GQA.

• Self-motivated learner who stays current with ML research and contributes to team knowledge sharing.

What We Value

• Agency – You move decisively through ambiguity and take full ownership.

• Taste – You have a sharp eye for quality across research investments, product, and engineering.

• Craft Obsession – You’re energised by solving hard problems and continuously levelling up your work.

Why Join Us?

• Career growth – Join early and grow with us in a fast-scaling, high-impact environment.

• Direct impact – Work closely with founders, legal experts, and engineers to shape the future of legal AI.

• Equity – Share in our success with meaningful ownership in the company.

• Purposeful work – Build technology that transforms how the legal industry works from the ground up.