AI Architect

The Company

A well-funded, Series A Australian startup focused on improving application performance through advanced computational waste detection.

Strong global backing, early traction, and a clear technical vision. The team is tackling hard, meaningful problems at the intersection of systems, AI, and performance.

The Role, and what you’ll be doing

This is a rare opportunity to build a world-first capability in neural network and LLM optimisation.

You’ll identify high-impact opportunities and translate them into production-ready innovations; improving inference performance, model efficiency, and cost at scale.

You’ll operate across research and engineering; evaluating cutting-edge techniques and ensuring they ship into real systems.

Key ResponsibilitiesInference & Compute Optimisation - Design and implement highly optimised inference pipelines and computational kernels; leveraging SIMD vectorisation, cache-aware memory access, and hardware-specific tuning.

Neural Network & Model Optimisation - Apply pruning, quantisation, and compression techniques across LLM and vision models; balancing performance gains with model quality.

Profiling & Observability - Build and use advanced profiling tools to identify bottlenecks across the stack; from memory and compute to end-to-end pipeline performance.

Evaluation & Benchmarking - Develop rigorous benchmarking frameworks; enabling systematic comparison across optimisation strategies and measuring real-world impact.

Technical Leadership - Act as a senior technical leader; mentoring engineers and driving a culture of high-performance, research-to-production execution.

About You

• Deep systems expertise; 8+ years in high-performance computing, AI systems, or low-level optimisation
• Strong understanding of CPU/GPU architecture, memory hierarchies, and performance tuning
• Proven track record optimising neural networks and LLMs; from research through to production
• Experience with pruning, quantisation, and inference acceleration techniques
• Comfortable working across research and engineering; translating theory into shipped systems
• Strong communication; able to align engineering and research teams

Nice to have

• Experience with C/C++, inference engines, or x86 intrinsics
• Background in neural network compression or applied AI research
• Publications or patents in AI/ML optimisation

Why this role

• Work on genuinely hard, high-impact problems
• Build world-first optimisation capabilities
• Join a well-funded team with strong technical ambition
• Operate at the intersection of AI, systems, and performance

If this sounds like you, or you’re curious to learn more, let’s have a confidential conversation.