Most software naming is cosmetic. You call your message queue Kafka and your cache Redis — the names describe nothing about how they should evolve. When your cache needs a new feature, the name gives you zero guidance.
We tried something different: using cell biology as an engineering manual for an AI agent orchestration system. Not metaphor — constraint. The names do not just label components, they predict what features to add, what to remove, and how pieces should interact.
mtor dispatches AI coding agents to execute tasks. Ribosome is the agent that takes a spec and produces code. Named after their biological counterparts: mTOR, the mechanistic Target of Rapamycin, is the cell’s master regulator that integrates nutrient signals and decides whether to build proteins. The ribosome is the molecular machine that reads mRNA instructions and assembles proteins. The naming was not arbitrary. We chose it because the mechanisms matched: mTOR decides what to build and when, ribosomes execute. But we did not anticipate how much the biology would teach us.
During an overnight autonomous dispatch session, we hit three problems. Each time, the biology had already solved it. Tasks completed successfully but the review gate could not see their output — we were checking one git diff range that was structurally empty when the agent committed directly to main. Real mTOR does not just check one signal. It integrates multiple pathways for amino acids, growth factors, and energy. So we built a fallback: record HEAD before execution and use that as the comparison base when the primary range is empty. Multiple sensing pathways for the same signal.
The review gate rejected seven tasks. Manual investigation revealed all seven had produced working, tested code. The work existed but was invisible to the gate. When nutrients are scarce, mTOR inhibition triggers autophagy — the cell literally eats its own components to recycle valuable materials. Nothing is wasted. So we designed a salvage command: scan rejected tasks, check if the build server has commits that the dispatch server has not merged, recover them automatically. The biology said do not discard rejected work, recycle it.
We dispatched at a constant rate regardless of outcome quality. When the review gate was broken and rejecting everything, we kept dispatching into a broken pipeline. In biology, mTOR has a negative feedback loop: when ribosomes produce misfolded proteins, a phosphorylation cascade feeds back to reduce mTOR activity. The cell slows down protein production when quality is poor. So we designed rejection rate tracking. When over sixty percent of recent tasks are rejected, halve the dispatch rate and double the check interval. When quality recovers, restore the original rate.
Equally powerful: the biology told us what to remove. We had a numeric satisfaction score from zero to a hundred with deductions and bonuses. But mTOR phosphorylates or does not — it is binary. There is no seventy percent phosphorylated. The score actively confused debugging. Flags like no_commit_on_success were the real diagnostic all along. We killed the score. We had built an active probe command that pings providers to check health. But mTOR senses nutrients passively from metabolic byproducts — amino acids activate pathways as a side effect of normal metabolism, not through dedicated probes. Our circuit breaker already updated passively when tasks completed. The probe was un-biological overhead. We removed it.
The common objection is that these are just fancy names for circuit breakers and recycling. Yes — but the names generated those designs. We did not start with “we need a circuit breaker” and then name it. We started with mTOR and asked what does real mTOR do that our system does not. The biology is a checklist of solved problems. Eight features. Zero were designed from requirements. All were read from biology.
Single metaphors exhaust quickly. “Pipeline” gives you input, process, output, and then you are done. Cell biology has 3.5 billion years of solved problems at every level of abstraction. The naming convention does not just label — it generates an infinite roadmap of features that are already proven to work together, because the cell already integrated them. The test is not whether the name is clever. The test is whether the name tells you what to build next. If it does, it is not a metaphor. It is an engineering manual.