SYS / SHOWCASE-EU-1
DA
[ system / architecture ]· synthetic demo data

Layered by responsibility.
Complexity is earned.

Each layer exists for a specific reason. Select a layer to see its purpose, why it exists and what it depends on.

[ system layers ]
[ layer 06 ]

Execution Strategy Layer

Purpose

Chooses how much autonomy a task gets.

Why it exists

Complexity is earned: more agents are used only when the task, risk and governance needs justify them.

Dependencies

Model provider abstraction, retrieval

Deterministic
Why chosen

For straightforward or rules-based cases it can be cheaper, faster, easier to audit and safer. A first-class path, not a lower-quality fallback.

Trade-off

Less adaptive reasoning than agentic execution.

Single agent
Why chosen

One bounded reasoning loop with selective access to approved knowledge tools, when tool choice improves context gathering.

Trade-off

Less separation of responsibilities, but lower coordination overhead, latency and synthesis risk.

Multi-agent
Why chosen

Separates evidence, architecture, risk/governance and synthesis when independent specialist analysis is justified; can surface meaningful disagreement.

Trade-off

More latency, cost, failure surfaces and synthesis risk. Not inherently superior — complexity is earned, not assumed.

Provider abstraction
Why chosen

Separate contracts for structured generation and embeddings, so changing the reasoning provider does not force re-embedding. OpenAI: structured generation + embeddings. Anthropic: structured reasoning/tool selection, not embeddings.

Trade-off

More abstraction code, less provider coupling.

Failure behavior

Explicit timeout and retry rules. Permanent validation failures are not retried; timeout, connection, rate-limit and selected 5xx failures may be retried within limits. Safe failure categories are returned instead of raw provider exceptions.

Confirmed configuration →