Founder / NeuroforgeIO Pty Ltd

Lloyd Handyside

Lloyd Handyside founded NeuroForge to build expandable cognitive systems and show what becomes possible when AI amplifies multidisciplinary human expertise.

Founder briefing

Why NeuroForge exists—and what comes next.

Lloyd explains the ERAIS thesis, the two paths into the architecture, and the evidence discipline behind the research.

1 min 34 secEnglish captionsTranscript below

Accessible alternativeVideo transcript

Download plain text

Hi, I'm Lloyd Handyside, founder of NeuroForge.

I started ERAIS from a fairly simple question. Why do we treat brute-force compute and throwing away already-trained models as inevitable parts of advancing AI?

ERAIS, the Efficient Routed Adaptive Inference System, E-R-A-I-S, is a novel sparse and adaptive AI architecture designed to use computation selectively rather than activating everything uniformly.

We're developing two paths into the same architecture. New models can be trained natively as ERAIS, and the second path is to convert existing dense models into ERAIS, with the goal of preserving useful capability while reducing the active compute needed to run, update, and continue training them.

Supporting that architecture is a rigorous research and analysis stack: exact checkpoint recovery, model and data lineage, controlled evaluation, the ERAIS Observatory, and evidence controls that keep measured results separate from projections and unproven claims.

We're already seeing strong, bounded engineering results across CPU and GPU systems, but the bigger point is how we do the research. Negative results stay in the record, and claims move forward only when the evidence does.

The goal is straightforward: make advanced AI more efficient, more accessible, and easier to keep improving without throwing away what already works.

That's what we're building at NeuroForge.

On this page

I built NeuroForge around a simple idea: use the right tools and effort for the task. Do not make the whole system work when only part of it is needed.

That idea became ERAIS. It also became the way I run the company.

My background cuts across biology, physics, chemistry, big data, programming, industrial environments and academic research exposure.

I do not keep those disciplines in separate boxes. I bring them together and use whichever view helps me understand the problem in front of me.

I left my previous work to build NeuroForge full time. The company is also a live test of a practical question: what pace of research and execution becomes possible when disciplined human judgement is deeply amplified by AI?

That is part of why ERAIS does not look like another language-model project. I did not assume that today’s dominant AI architecture must be the final shape of machine cognition.

I started with a more basic question: if we define cognition from first principles, what should a computational cognitive system actually look like?

AI should amplify people, not erase them

NeuroForge has been AI-native from the beginning. That does not mean people are obsolete.

I use AI to hold context, explore alternatives, implement, test, analyse, research and iterate at a pace I could not sustain unaided.

I remain responsible for the direction, synthesis, judgement, experiment design, company decisions and definition of what matters.

That combination has allowed one founder to advance a substantial research system while also handling company formation, commercial product design, investor work, market research, research discussions, pilot qualification, public infrastructure and evidence control.

The point is not that one person should do everything forever.

The point is that a capable person working deeply with AI can operate with far greater breadth and iteration speed.

As NeuroForge grows, I want to add amplified experts: researchers, engineers and operators, each working with AI inside one coherent evidence and decision system.

Why I think this matters

The NeuroForge operating loop is a form of human-in-the-loop recursive improvement.

Current AI helps me build the next cognitive system. Better systems should help us do the next round of work faster and better. That means more experiments, more data, better tools and quicker architectural learning.

Each useful result improves the next iteration.

The human stays in the loop. Scientific discipline stays in the loop. Reality stays in the loop.

ERAIS makes that loop especially interesting. Today’s AI is helping build a proposed successor architecture designed to learn, expand and use computation more selectively than the systems helping to create it.

That is the programme. The evidence still decides how far the thesis earns the right to go.

The founder principle

I do the impossible because most things called impossible are just incorrectly defined.

That does not mean ignoring constraints. It means finding the real constraint.

A problem can look impossible because it has been framed at the wrong level. Change the representation. Remove an assumption. Isolate the real bottleneck. Stop solving the wrong problem.

That is how I approach research. It is how I approach the company. And it is the philosophy behind ERAIS itself.

What I am building

NeuroForge is building four connected things:

an expandable cognitive architecture;
the tools and evidence systems needed to develop it responsibly;
a commercial path that applies the same philosophy to real customer problems; and
an AI-amplified organisation designed to learn faster as it grows.

The ambition is not small: intelligence that can grow without making every thought proportionally more expensive.

Read ERAIS, the company story, the evidence, or the funding and partnership path.

Current stage

NeuroforgeIO Pty Ltd is incorporated in Australia under ACN 701 314 329 and ABN 87 701 314 329.

The company is founder-led and pre-revenue. Current technical claims, commercial status and external relationships are recorded in the public evidence and company-state records. They are not inferred from the ambition described above.