Security Objectives / deeper technical view
Security Objectives builds tools for inspecting computational state, tracing AI-generated information and exploring controlled AI infrastructure. RootCompute brings those evidence-producing systems into one broader architecture.
Our portfolio
Runtime integrity and AI provenance form the core of the portfolio. Each addresses a different evidence problem; neither substitutes for the other.
01 / Computational state
A composable integrity engine for bare-metal, virtual, mixed host/hypervisor, nested page-table and SmartNIC-connected topologies. Reconstruct memory context and validate executable code within the pipeline you choose.
02 / Generated information
A research toolkit for watermarking and provenance in generated language. Bring generation, detection experiments and inspectable artifacts into one workflow.
From the lab
Alpha / in development
Your local inference fleet.
An open-source foundation for bringing multiple Ollama instances behind one endpoint. Planned work includes cross-host layer handoff, capability-aware placement and predictable failure behavior; those remain forward-looking directions.
Explore ololComing soon
A closer view of agent-assisted work.
A forthcoming toolkit focused on session visibility, recording and diagnostic context for AI-assisted workflows. Supported workflows and availability will be shared as the product takes shape.
Ask about MCP-PTZinVtero.net / deployment breadth
The inspection path and the system being inspected are different parts of the architecture. Keep both visible.
inVtero.net / Nested
Recursive introspection accounts for nested page-table and hypervisor relationships, rather than treating every address as if it belonged to one flat machine.
This is a logical view of nesting, not a claim that every platform exposes the same hardware page-table walk.
Observation
Recursive translation context
Reconstruct
Verify
Interpret
XAMMY / from generation to evidence
A watermark is one signal. A retained record provides different evidence. XAMMY is designed to evaluate those pieces separately and preserve the context needed to interpret them.
XAMMY / Generate
XAMMY explores watermarking at generation time. A configured sampler shapes token choices while the reader sees ordinary text.
MLX on Apple Silicon · vLLM on NVIDIA CUDA
Our engineers found that the proposed design reduces latency without increasing cost.
AI direction / inVtero roadmap
The roadmap is to make reconstructed state and integrity results usable by AI-assisted security systems while keeping observations, verification results and model interpretations distinct.
01
Keep the observation source, authorization and access boundaries explicit.
02
Recover supported address spaces, executable pages and topology context.
03
Compare observations with reference material and preserve the result.
04
Use evidence to guide investigation and policy without turning inference into fact.
Design partners & technical collaboration