How a production AI system is put together
Most of a working AI system is not the model. It is the workflow that controls execution, the tools that reach real business systems, and the data and observability layer that makes the result auditable afterwards.
- Research agent
- Analysis agent
- Validation agent
- Specialist agent
- Analyst agent
- Reviewer agent
- Action agent
- Integration agent
- Verification agent
What each layer is responsible for
And the failure each one prevents when it is present, or causes when it is not.
| Layer | Responsibility | Failure it prevents |
|---|---|---|
| Application | Delivers the experience: web, mobile, SaaS or API. Holds identity, permissions and the user-facing contract. | AI capability with no way for a business to actually use it |
| Flow orchestration | Sequencing, durable state, conditional routing, retries, validation and human approval. Written as deterministic code. | Workflow logic hidden in a prompt, where it cannot be tested or reasoned about |
| Crews | Groups of specialised agents that collaborate on one part of the problem, under the orchestrator. | One oversized agent trying to hold an entire process in a single context |
| Agents | A defined role, an explicit tool grant, a typed output contract, a timeout and a documented fallback. | An agent that can take an action nobody authorised |
| Tools and integrations | Every external effect: APIs, CRM and ERP, databases, documents, search. Permissioned and logged. | A model that appears to act but changes nothing real |
| Model layer | Language, embedding and vision models behind an adapter, selected per task and cost profile. | Vendor lock-in, and a rewrite every time a model is replaced |
| Data and observability | Application database, vector store where needed, workflow state, logs, tracing and an append-only audit trail. | A production incident nobody can reconstruct |
Decisions we apply on every build
Six rules, each visible in the public code or a live system on this site.
The model reads and drafts. It does not decide.
Extraction, classification and drafting are language work. Scoring against a rule, computing a figure or committing an irreversible action belongs in deterministic code.
Withholding beats guessing
A field the system is not confident about is returned empty and flagged, never filled with a plausible value. In retrieval, a relevance floor decides whether an answer is composed at all.
Human approval is a workflow state
Not a convention in a prompt. The run pauses, persists, and resumes on a person’s decision, with the approval recorded.
Accuracy is measured, not asserted
A held set of real cases, scored on every change, with precision and recall reported separately rather than as a single number.
Providers sit behind an adapter
Changing model or vendor is a configuration decision. One system we built runs two providers on a single orchestration layer.
Every run is reconstructable
Inputs, tool calls, latency and cost per execution. If it cannot be replayed, it cannot be debugged at two in the morning.
The same rules, in code you can read
Our public retrieval system applies these decisions end to end: a grounding threshold that stops generation, filters applied inside the query, agents with documented fallbacks, and citations verified against live records.
Bring us your architecture
Sixty minutes on the system you have or the one you are planning, and an honest read on where AI belongs in it.