A$695k pre-seed / Queensland, Australia
Today's AI often gets more capable by making the whole system larger and more expensive to run. NeuroForge is building ERAIS to test another path: add new capability, then use only the parts each task needs.
NeuroForge is seeking A$695,000 for an 18-month plan. The plan assumes no customer, grant, partner, debt or founder-cash receipts. Public information only; not an offer to issue securities.
What the company is building
ERAIS is the long-term architecture. Fracture explores whether existing learned capability can be brought across. The 10-Day AI Fit Test offers a way to work on present-day customer problems without asking buyers to depend on unfinished ERAIS research.
01 / Architecture
ERAIS is designed to store many capabilities while applying only the appropriate tools and effort to each task. The next question is whether that advantage holds as the system grows.
02 / Transition
Fracture explores whether capability learned by conventional models can be converted into ERAIS structures. A bounded end-to-end research workflow already runs.
03 / The company
NeuroForge combines human judgement, capable AI tools, clear coordination and retained evidence. Funding adds specialist expertise without losing the way of working that created the system.
Looking to support a smaller project? Watch Minecraft AI live and choose voluntary monthly support. This is separate from investment and does not buy shares or game access.
An incorporated company. A working ERAIS research system. Public evidence controls. A defined commercial offer. A live technical-diligence site.
Long native-byte training runs have completed. Bounded Fracture, multimodal and governed paths run. Quality and scale remain live research questions.
Challenge 77 application submitted and pending. A CQU research proposal remains in draft for NeuroForge review.
No customer, partner, award, grant, investor, paid pilot or independent technical validation is claimed today.
Why the lead can keep growing
To catch up, a competitor would need the architecture, the conversion knowledge, the record of what worked and failed, the research tools, and the way NeuroForge puts those parts together.
Knowledge of selective execution, modular capacity, recurrence, memory and multiple kinds of input.
Knowing what survives conversion, what fails, and what needs to be repaired.
Accepted results, negative results and decisions that do not need to be rediscovered.
Systems for training, testing, tracking evidence, finding faults and restoring known-good states.
Human judgement and AI tools coordinated inside one disciplined research process.
Biological, physical, computational and industrial insight that can eventually inform new hardware.
Each useful result informs the next experiment. A competitor copying one snapshot would still need to reproduce the research record and the way it was built.
Why Lloyd / why this model
Lloyd Handyside draws on biology, physics, chemistry, big data, programming, industrial work and academic research exposure.
That range of experience shaped ERAIS. AI turns one founder's time into far more research and engineering iteration—without handing judgement to the tools.
Funding is not meant to preserve a one-person company. It adds researchers, engineers and operators who can use the same method inside one coherent scientific system.
Read the founder storyUse of funds
The plan stays lean. It pays for hardware and AI tools, independent validation, pilot delivery, essential business support, milestone specialists, and one support engineer only when field work creates the need.
What 18 months must deliver
The base case assumes no customer, grant, programme, research-partner, tax-incentive, debt or founder-cash receipts. Capital buys the time and evidence needed to make a clear next-financing or stop decision.
The 10-Day AI Fit Test and staged deployment path turn the NeuroForge method into paid customer decisions now. ERAIS remains the architecture that determines the company's largest possible future.
An independent technical conclusion on the largest remaining ERAIS risk.
A clear quality, resource and continuation decision for the next native ERAIS stage.
At least one funded customer engagement with witnessed evidence and a real buyer decision.
A customer-operated release with security, recovery, second-operator and support boundaries.
A documented decision to raise the next round, continue from revenue, redesign or stop.
Capital and funded partnerships
Capital funds the decisive experiments. The right introductions open conversations with buyers, research partners and independent technical reviewers.