Jylus Systems Pty Ltd
Jylus is a governed evidence and state layer for enterprise AI. It ingests live data, resolves current and historical state and relationships, detects conflicts and missing evidence, and compiles source-backed Context Packs for models and agents.
At a Glance
- Enterprise AI teams
- Developers and AI-agent builders
- Security and observability teams
- Operational and incident-response teams
- +2 more
AI Tools by Jylus Systems Pty Ltd
(1)
Jylus
Governed Evidence Layer for AI
Discussions
No discussions yet
Be the first to start a discussion about Jylus Systems Pty Ltd
Latest News
Jylus published its core subprocessor list, naming OVHcloud, MongoDB Atlas, Mailjet and Stripe.
JYLUS trademark filed for cloud-based SaaS hosting services.
Jylus Service Terms became effective for account and service use.
Jylus Privacy Notice became effective, documenting handling of account, billing, security and customer data.
Products & Services
A model-independent API that ingests live data and provides source-backed evidence. The documented production operations are event ingest, search, exact query and analyze/evidence preparation.
Compiles compact, proof-bound context from current state, history, relationships, semantic matches, contradictions and missing-evidence signals for use by any AI model.
HTTPS event ingestion by default, with managed NATS JetStream for sustained or high-throughput deployments.
Downloadable Docker log and OpenTelemetry collector deployments that buffer and deliver data through the tenant-scoped HTTPS ingest path.
Market Position
Jylus positions itself as an evidence and state layer used alongside an existing data stack and model, rather than as a replacement model. Against conventional vector search/RAG, it emphasizes as-of state resolution, temporal history, relationship traversal, contradiction and missing-evidence detection, and proof IDs for every retained claim. The company reports 528/528 observed strict accuracy, 98.77% less model input and 23.6x faster evidence preparation on its defined workloads, while explicitly limiting those results to the published benchmarks.
Leadership
Founders
Josh Hodgetts
Founder and CEO of Jylus; the company says it was founded in Perth. No prior roles or companies were identified in the reviewed authoritative sources.
Executive Team
Josh Hodgetts
Founder & CEO
Founder and chief executive of Jylus; identified by the company as the Perth-based founder.
Founding Story
Jylus was founded in Perth by Josh Hodgetts to address the problem of AI systems reasoning over stale, conflicting or irrelevant data. Its initial vision is to establish what was true, when it was true, and whether the available evidence is sufficient to answer, by placing a proof-bound evidence layer between customer data and any model.
Business Model
Revenue Model
Subscription SaaS priced by workspace capacity, event throughput, searchable storage/history, concurrent queries and support level. Enterprise capacity, retention and service terms are custom; paid plans renew for the selected billing period.
Pricing Tiers
1,000 events/sec, 1 GB searchable storage, 7 days searchable history, 4 concurrent queries and HTTPS API access; no credit card required.
2,000 events/sec, 5 GB searchable storage, 14 days searchable history, 8 concurrent queries, managed HTTPS and NATS.
5,000 events/sec, 10 GB searchable storage, 30 days searchable history, 16 concurrent queries, managed HTTPS and NATS.
100,000 events/sec, 250 GB searchable storage, 365 days searchable history, 256 concurrent queries, managed HTTPS and NATS.
Custom capacity and retention, architecture review, named support and contract-defined service terms.
Target Markets
- Enterprise AI teams
- Developers and AI-agent builders
- Security and observability teams
- Operational and incident-response teams
- Prediction and telemetry workloads
- Side projects, early-production applications and startups
- AI models and agents that need compact, source-backed context
- Security, observability, telemetry and incident investigation
- Operational decisions requiring current-versus-historical evidence
- Prediction systems using temporal patterns, relationships and historical matches
- Enterprise data ingest, retrieval and evidence preparation