Building the infrastructure for the next generation of AI

We're on a mission to democratise access to powerful AI infrastructure, giving organisations full control over their assets, deployments with the privacy & sovereignty of their data.

Our Mission

TensorFoundry is dedicated to democratising access to powerful AI infrastructure. We believe that organisations should have full control over their AI deployments, with the privacy, performance and cost benefits that come from edge and on-premises infrastructure.

Our products enable developers and enterprises to deploy state-of-the-art LLMs anywhere - from edge devices to data centres - without compromising on performance, security or sovereignty.

Our approach to consulting begins with data - the most valuable asset any organisation holds. We focus on understanding and shaping data before applying AI or automation, ensuring every solution is grounded in real insight and knowledge.

Our Values

Privacy First

We believe data privacy is a fundamental right. Our infrastructure keeps your data on your hardware, under your control.

Open Source

We're committed to open source. Olla and key components of our stack are freely available for the community.

Performance

We obsess over performance. Every optimisation, every feature is designed to maximise efficiency and minimise latency.

Developer Experience

Great tools should be a joy to use. We prioritize clear documentation, intuitive APIs and helpful error messages.

Meet the Founder

Thushan Fernando

Founder & Principal Engineer

Melbourne, Australia

Thushan is the founder of TensorFoundry and the lead engineer behind Olla, the open-source LLM proxy and load balancer. Over 25+ years building software, he's specialised in low-level, performance-critical systems work across search, machine learning, and GPU-heavy compute, mostly in Go, Rust and Zig. Previously at SEEK and Readify, where he worked on low-latency search infrastructure and built NeoSearch, an efficient edge-model search platform serving ML at scale. TensorFoundry pivoted to AI infrastructure when the first Llama models landed; the bet was that on-premise inference would matter as much as the models themselves. Based in Melbourne, Australia.

Low-latency systems
Search & information retrieval
Machine learning infrastructure
GPU compute
Go
Rust
Zig

Our Journey

From early experiments with Sherpa in 2018 to pioneering edge AI infrastructure with FoundryOS, we've been on a mission to democratise powerful AI deployments. Our journey has been driven by real-world needs and a commitment to making AI accessible for organisations of all sizes.

Starting with Scout's predictive analytics, evolving through Olla's open-source success with over 20K+ downloads and now building FoundryOS for enterprise scale - every step has been informed by a decade of hands-on AI infrastructure experience.

Explore Our Full Journey

2018

2023

2025

2026