Analytics
Inputless analytics for operational decisions
Keeps reading your data, flags odd patterns, and suggests a next step. One picture for the desk.
Operation as a Service
Reservedanalysisforhigh-stakesdecisions
We assess your case, deliver the findings, and a Decision Report you can defend to leadership and the board.
- Every recommendation traces back to the source data.
- Your data stays on your network.
- One final report, ready for the board.
Engagement
From evaluation to a decision report
We operate as Managed AI Service: Operation as a Service. You get results after case evaluation, then a complete Decision Report that captures the full decision trail.
The pipeline below is how we move from intake to a shareable PDF, no hand-wavy “phase gates,” just a clear sequence you can plan around.
- Step 01
Case evaluation
Scope & fit
We assess context, constraints, and decision stakes so every engagement starts from a shared understanding of what “good” looks like.
- Step 02
Results shipped
Operations first
Operation as a Service means tangible outputs from the evaluation (clarity, findings, and directional results) before anything is wrapped in narrative.
- Step 03
Decision Report
Complete synthesis
We then produce a full Decision Report: rationale, trade-offs, options, and recommended authority, so decisions are defensible, not improvised.
- Step 04
PDF deliverable
Ready to share
The same report is delivered as a polished PDF for leadership, boards, and partners: one artifact from evaluation through final decision record.
Step 01
Case evaluation
Scope & fit
We assess context, constraints, and decision stakes so every engagement starts from a shared understanding of what “good” looks like.
Step 02
Results shipped
Operations first
Operation as a Service means tangible outputs from the evaluation (clarity, findings, and directional results) before anything is wrapped in narrative.
Step 03
Decision Report
Complete synthesis
We then produce a full Decision Report: rationale, trade-offs, options, and recommended authority, so decisions are defensible, not improvised.
Step 04
PDF deliverable
Ready to share
The same report is delivered as a polished PDF for leadership, boards, and partners: one artifact from evaluation through final decision record.
End state
One continuous thread: evaluate→ship results→Decision Report→PDF in your hands
Example
See a real Strategic Decision Report (controlled mushroom production, multi-room sensors, predictive cultivation intelligence) and the case analysis on Case Studies.
Open sample PDFHow we work
From a single case to ongoing analysis
We read the context, the constraints, and what is at stake. You leave with directional findings and a Decision Report written for leadership and the board.
Request a case evaluationWhat the work stands on
Judgment you can defend
Traceable
Every recommendation walks back to the source data
Your network
The data does not leave your perimeter
One report
A single Decision Report, ready for leadership and the board
Cybersecurity
Advanced protection paired with the analysis
Where we operate
Intelligence for desks where delay is expensive
Energy
Planetary-scale operations and grid intelligence
Model generation, transmission, and market signals in one continuous view so teams see imbalance and risk before they cascade.
Energy01 / 05
The deliverable
A report a board can use
The Decision Report sets out findings, scenarios, risks, and options. Each recommendation traces back to the source data. The work stays on your network.
What you receive
- Context, constraints, and what is at stake
- Findings you can defend in a briefing
- Scenarios, risks, and options
- A trail from each recommendation to the source data
When it is the right instrument
Energy, cybersecurity, defence, healthcare, and utilities: decisions where a wrong call is expensive and the reasoning has to hold in front of a board. The useful output is a judgment, not another dashboard.
Request a case evaluationMethod and research
Internal research programs
Not for sale
The labs behind the analysis
Our research programs (Myotis, Alcatoe, Moneyless) are the internal labs where we develop the methods used in client work.
They are not products. They make each analysis faster, deeper, and traceable back to the source data.
Myotis
Internal lab for legal and obligation analysis: norms, duties, and facts, so contractual risk is visible before a renewal or a dispute.
Discover the methodAlcatoeInternal lab for defence and cognitive security: one picture of tasking, logistics, and indicators, with the reasoning still inspectable.
Discover the methodMoneylessInternal lab for payments and fraud: large flows watched without lighting up every transaction, and without a noisy alert pile.
Discover the methodCase histories
Our case histories, your success stories
Analytics, security, defence, and payments on the same stack. Your data stays where it is.
Cybersecurity
Security that includes how people are misled
Alcatoe helps a SOC catch disinformation, map how events connect, and explain the call, not only the alert.
Defence
Wargames and simulators, not slide decks
Exercises and multi-domain simulators for deception, speed, and the messy links between cyber, physical, and information operations.
Financial services
Payment routing that learns as it goes
Moneyless picks a provider, watches for fraud, and updates itself. Routing is not a static rule spreadsheet.
Analytics
Inputless analytics for operational decisions
Keeps reading your data, flags odd patterns, and suggests a next step. One picture for the desk.
Cybersecurity
Security that includes how people are misled
Alcatoe helps a SOC catch disinformation, map how events connect, and explain the call, not only the alert.
Defence
Wargames and simulators, not slide decks
Exercises and multi-domain simulators for deception, speed, and the messy links between cyber, physical, and information operations.
Financial services
Payment routing that learns as it goes
Moneyless picks a provider, watches for fraud, and updates itself. Routing is not a static rule spreadsheet.
Explainable AI
Why did the AI decide this?
Black-box scores do not survive a briefing. The XAI path walks a recommendation from the source record to a short narrative, so a lead can still own the call and show how they got there.
Trace the chain
Justified output
Strategic narrative + audit path
Source item
Structured feed record (e.g. GDELT event line or radar track update) ingested with timestamp and provenance.
gdelt:gcam_v1 · evt_hash 8f3a… · src CONFIRMED
Every stage remains inspectable: from the originating sensor or feed line to the wording your command staff sees. No unexplained jumps.
GRID REF // CYBER 01
Offensive services
Cybersecurity
Authorized work, a written scope, findings you can act on
Alongside the analysis, Talorik delivers offensive cybersecurity: adversary-led assessment, intelligence, and forensics, inside a written scope and rules of engagement.
Cognitive Security (COGSEC) is the work of studying and defending against attacks on judgment and cohesion. We do that with Inputless Alcatoe and with wargames that treat the opponent as someone who thinks, not a script in a binder.
Offensive services
Offensive Security
Authorized assessment of the attack surface, inside written rules of engagement.
Read moreThreat IntelligenceActors, campaigns, and indicators read against your environment.
Read morePenetration TestingTime-boxed tests of agreed targets, with reproducible findings.
Read moreRed Team OperationsA defined mission against people, process, and technology.
Read moreAdvanced Malware AnalysisReverse engineering in an isolated lab, with indicators the SOC can use.
Read moreZero-day ResearchGoverned research on systems you are licensed to test.
Read moreAPT SimulationA named actor’s tradecraft, measured against your controls.
Read moreForensic ServicesEvidence, a timeline, and a report that holds after an incident.
Read moreArchitecture
How the stack is put together
How the pieces fit
From feeds to a call the desk can defend
Classic analytics assume you already know what to look for. You write a query, wait for a chart, and hope it still matters. Talorik keeps reading the feeds and flags what looks off, so you are not stuck guessing the question.
Inputless Analytics is the name for that setup: data enters Talorik OS, goes through the ontology, the kinetic graph, and your own security boundary, and comes out as a recommendation someone can defend.
Problem · You have to know what to ask
Most analytics wait for someone to write the right question. If you miss that question, you miss the event. Dashboards catch up after the fact, and each system tells a slightly different story.
Solution · Inputless Analytics
The system keeps reading your feeds and points at what looks off. The Cognitive SDK watches the stream, the Kinetic Graph maps how things connect, and you get a suggested next step instead of another blank search box.
Signal path · From feeds to a call
Operational data goes into Talorik OS. The model sorts noise from signal and hands you a recommendation you can check against the original source.
Raw DataTalorik OSDecisions
Architecture · The Ontology
A live model of how your organization actually reasons: what an event means here, what people usually do next, and the extra types you need for your domain.
Architecture · The Kinetic Graph
Relationships between entities, not rows in a table. Twenty-two node types, thirty-plus relationship types, stored in Inputless DB. You see cause, not just that two numbers moved together.
Architecture · Data stays put
Processing happens inside your network. The model does not need to leave. PII is kept to a minimum. GDPR is a design constraint from the start.
GRID REF // AGENT 01
AI Agents
Capability
Agents that watch, reason, and suggest a next step
Agents here are workers with perception, memory, and a policy engine. They watch operational feeds and rank what deserves a look before anyone opens a query.
Built on the Cognitive SDK and the Substrate ontology, they turn live feeds into Courses of Action you can trace from signal to suggestion, under your rules.
Perception · Continuous sensing
Agents read operational feeds, logs, and signals without waiting for someone to frame the question. Context builds up before anyone asks.
Read moreReasoning · Inside your rulesA local reasoner with memory and a policy engine ranks options and keeps every suggestion inside the rules you set.
Read moreAction · Courses of ActionYou get ranked Courses of Action (COA). Leadership picks a direction. Nobody is babysitting discovery across six screens.
Read moreWhere they run · On your networkAgents run where the data lives. Nothing has to leave the building for the system to think.
Read moreNewsroom
News and ideas
Notes on the stack, the graph, and running the model on your network.
All articlesGRID REF // RSC-01
Papers
Papers and notes. One read, no extra chrome.
Partners
Trusted by leading firms
Studio Errico
Criminal
Dendrotic AI
Studio Errico
Criminal
Dendrotic AI