
The Jenny Manifesto-A New Class of Architectural Intelligence for the Age of Learning Machines
- Mark Kendall
- 6 days ago
- 5 min read
The Jenny Manifesto
A New Class of Architectural Intelligence for the Age of Learning Machines
1. The Problem Nobody Is Naming
Every enterprise today is building software faster than it can understand it.
We have:
Microservices.
Cloud platforms.
CI/CD pipelines.
Infrastructure as code.
Hundreds of repositories.
Dozens of teams.
Thousands of configuration files.
And now… AI agents everywhere.
And yet, one brutal truth remains:
Nobody actually knows what their system is becoming.
Architectural intent is written once (if at all) and forgotten.
Standards are published and immediately violated.
Diagrams rot.
ADRs go stale.
Governance becomes a quarterly theater exercise.
Meanwhile, reality drifts.
Quietly.
Continuously.
Unnoticed.
Until one day leadership asks:
“Why is delivery slowing down?”
“Why is everything so brittle?”
“Why can’t we change anything safely anymore?”
“How did we get here?”
And nobody has a real answer.
2. The Coming Agent Chaos
Now we are adding AI agents into this already fragile system.
Copilots writing code.
Bots generating pipelines.
Agents opening PRs.
LLMs modifying configs.
Automation changing infrastructure.
Each one is locally “smart.”
None of them are globally responsible.
We are creating a swarm of task bots without a governing brain.
So the real future problem is not:
“How do we get more agents?”
It is:
“Who is in charge of the behavior of all these machines?”
3. Jenny Exists Because Architecture Has No Immune System
Biological systems have immune systems.
Software systems do not.
There is nothing in a modern delivery platform that:
Continuously compares:
What we said the system should be
Versus what it is actually becoming
Detects architectural drift as it happens
Identifies slow-moving systemic rot
Raises early warning signals
Preserves institutional memory
Teaches the organization what it is doing to itself
So we built Jenny.
4. What Jenny Is (Precisely)
Jenny is:
A sovereign, context-native architectural governance agent that continuously reconciles declared intent with real system behavior and produces governance intelligence.
In plain English:
Jenny is an always-on Chief Architect that never sleeps, never forgets, and never gets tired of reading your repos, pipelines, policies, and standards.
She ingests:
Architectural intent
(YAML, Markdown, ADRs, standards, policies)
Code and repo structure
(Java, Helm, Terraform, pipelines, configs)
Runtime and infrastructure state
(cloud, Kubernetes, environments)
Delivery signals
(PRs, CI logs, failures, tickets, SLAs)
She reasons over:
“What did this organization say this system should be?”
“What is it actually becoming?”
“Where is it drifting?”
“Where are standards being violated?”
“Where is complexity accumulating?”
“Where are we creating future failure?”
She emits:
Drift reports
Governance violations
Architectural risk signals
Trend lines over time
Rot detection
Modernization and refactoring intelligence
Leadership-level architectural narratives
5. What Jenny Is Not
Jenny is not:
A chatbot
A Copilot
A code generator
A DevOps helper
A YAML explainer
A PR bot
Jenny does not exist to:
Make developers type faster
Generate boilerplate
Replace human judgment
Jenny exists to answer one question nobody else can answer:
“Is our system still becoming what we said it should be?”
6. The Category Nobody Else Is Building
Every AI product today is fighting over this layer:
“How do we help a human complete a task faster?”
Jenny lives in a different layer entirely:
“How do we govern the long-term evolution of a complex socio-technical system?”
She is not a tool-level agent.
She is:
A governance brain
A meaning engine
A drift intelligence system
A learning machine for architecture
This category does not yet exist commercially.
That is not a weakness.
That is the point.
7. Jenny’s Place in the Agent Stack
Jenny does not compete with Copilot, Duo, or pipeline bots.
She governs them.
┌────────────────────────────────────────────┐
│ Learn · Teach · Master (philosophy) │
└────────────────────────────────────────────┘
▲
│
┌────────────────────────────────────────────┐
│ Jenny — Architectural Intelligence Layer │
│ - intent vs reality reconciliation │
│ - drift detection │
│ - governance reasoning │
│ - modernization intelligence │
└────────────────────────────────────────────┘
▲
│
┌────────────────────────────────────────────┐
│ Tool-Level Agents (Copilot, Duo, Bots) │
│ - code generation │
│ - PRs │
│ - pipeline fixes │
└────────────────────────────────────────────┘
▲
│
┌────────────────────────────────────────────┐
│ Repos, Pipelines, Cloud, Infra, Tickets │
└────────────────────────────────────────────┘
Copilot writes code.
Duo explains pipelines.
Harness runs workflows.
Jenny decides whether the entire system is still architecturally sane.
8. Sovereignty Is Non-Negotiable
Jenny is sovereign by design.
That means:
She runs in your environment.
She uses your models if required.
She never exports your intellectual property.
She reasons over your standards, not generic best practices.
She learns your architecture, not the internet’s.
Jenny is not a SaaS hallucination engine.
She is an internal intelligence system.
9. Learn · Teach · Master Is Not a Slogan
It is the operating model.
Learn
Jenny continuously observes:
Drift
Violations
Patterns
Rot
Delivery friction
Governance failure modes
She builds a living knowledge graph of:
“How this organization actually builds software.”
Teach
Jenny produces:
Explanations
Reports
Trend narratives
Risk forecasts
Architecture health scores
She teaches:
Teams
Architects
Leadership
“Here is what your architecture is becoming.
Here is why it matters.
Here is what will break next if you do nothing.”
Master
Over time Jenny:
Refines intent specs
Tunes guardrails
Improves standards
Informs modernization strategy
Raises the architectural maturity of the entire org
She becomes:
The institutional memory and architectural conscience of the enterprise.
10. The Future Jenny Is Preparing You For
In 24–36 months, every serious enterprise will have:
Dozens of internal AI agents
Copilots in every IDE
Bots in every pipeline
Automation changing production daily
Leadership will finally ask:
“Who is governing all this behavior?”
The answer will either be:
“Nobody.”
(And chaos will follow.)
Or:
“Jenny.”
11. The Final Truth
We do not need more hype machines.
We do not need more task bots.
We do not need more dashboards that tell us what already broke.
We need learning machines that tell us:
“Here is what your system is becoming.
Here is why it matters.
Here is how to change course while it is still cheap.”
That is why Jenny exists.
12. One-Sentence Category Definition
Jenny is a sovereign, context-native governance agent that continuously reconciles architectural intent with system reality and teaches organizations how their software is actually evolving.

Comments