Proprietary capital · operating systems · integration in progress
Neurobrick operates independent proprietary-capital systems and is building a unified research, authority and evidence framework around them. Proposals, permission and execution remain distinct; end-to-end integration is still under verification.
The problem and the design response
Data, models, permissions and account state change on different clocks. Neurobrick’s design makes the boundary between each of them explicit, so a candidate can be questioned before it reaches capital.
A promising model result can still rely on stale data or exceed an account’s permitted risk.
Evaluate the research evidence and the applicable execution authority separately.
Tradeoff: More explicit handoffs; a high score alone cannot advance the system.
Research, inference and fleet tooling need broad visibility across systems.
Keep account context and execution permission inside the relevant engine boundary.
Tradeoff: Independent engines require additional coordination and reconciliation.
A deployed fix or successful process exit can leave the underlying operating problem unresolved.
Distinguish implementation, observed behavior and verified effect.
Tradeoff: Closure takes longer and remains limited to the measured scope.
These are architectural choices. Their enforcement and coverage must be established on each operating path.
Follow the decision path →Inspected operating footprint
Mesh registration: 8 devices, 6 online. These are network identities, not the physical operating-box count. A dedicated-fabric compute host is outside this mesh; inactive client devices are inside it.
Read-only inspection of the known operating fleet. Point-in-time internal disclosure, not live telemetry or independent attestation.
Development review · 14 September 2026
The late 14 September inspection found seven compute hosts, separate execution systems and operating support services. The unified path from research inputs to authorized execution, accounting and performance evidence remained under validation.
01 / Operating
Existing proprietary-capital execution systems, observation and infrastructure have operating evidence. This does not establish that every strategy is active or every integration is accepted.
02 / Under validation
Data provenance, evaluation, promotion controls and accounting are being connected and checked against deployed processes. A complete repeatable path is not yet established.
03 / Not established
This review does not establish durable net performance, readiness for automatic capital expansion or fleet-wide recovery. These require additional operational evidence.
Next acceptance conditions
Internal development review, inspected 23:30–23:32 KST. This is a scoped assessment, not live telemetry or an independent attestation.
A recorded stop · 14 September, 22:53 KST
Available data was not sufficient evidence. Related factors failed checks on source linkage, observation contracts and point-in-time provenance.
One saved readiness run and its specified input set. Counts are not a fleet-wide failure rate. This concerns the research observation stage and does not establish that all live trading was blocked.
Current operating capabilities · 14 September 2026
Runtime, authority and evidence signals are stated with their own scope. A running service does not by itself establish permission or verified performance.
Open system capabilitiesThree execution systems across separate hosts
Control plane HELD · release authority BLOCKED
22:53 KST run · lineage exit 2
NAV and margin artifacts reported OK
Local inference host · two running Linux CI VMs
System structure · separate logical roles
Research, market data and primary execution
A research and execution environment with domestic and US trader processes and separate order enforcement. Data readiness and strategy promotion are separate gates.
Independent domestic execution
A dedicated scheduled domestic rotation engine with its own execution and account boundary.
Multi-account execution
An account-scoped execution platform with scheduled workers and separate account contexts.
Control and evidence integration
A separate coordination layer supplying state, release controls and evidence services. It is not counted as a fourth independent execution system.
Claim boundary
A running system earns a stronger claim one verified path at a time.