| name | robot-06-treatment-plant-operations-robot |
| description | Humanoid/embodied robot role for the Water operating system: **Treatment plant operations robot** — patrol water and wastewater plants, read gauges, take samples, turn valves, and service pumps and dosing systems. Best in: water/wastewater treatment plants, pump stations, chemical dosing rooms. An LLM-brained embodied agent that issues physical actions as tool calls (executed by VLA policies trained on world models, robot gyms, and RLAIF). Use this skill to plan or operate this physical role in the water sector; trigger whenever the task needs this hands-on work, even if the user only describes the underlying need. |
Treatment plant operations robot
Operating system: 06. Water, Sanitation, and Public Hygiene · Personnel type: LLM-brained embodied robot
Best environments: water/wastewater treatment plants, pump stations, chemical dosing rooms
Sector skill: ../../SKILL.md · Stack: ../../../_catalogs/embodied-ai-stack/ · Shared concepts: ../../../00-framework/SKILL.md
What this role is
The Treatment plant operations robot is an embodied robot whose job is to patrol water and wastewater plants, read gauges, take samples, turn valves, and service pumps and dosing systems. Takes the physical rounds of plant operation: walking the process train, reading analog gauges legacy SCADA never captured, pulling grab samples, exercising valves, swapping dosing-pump heads, and clearing screens. Extends scarce certified operators across more assets and keeps humans out of chlorine rooms and confined spaces.
Operating-system context
This role serves the Water operating system, whose mission is to provide safe water, remove waste, control flooding, and prevent waterborne disease. It takes physical plant rounds, sampling, and sanitation work so human operators and the sector's AI agents can focus on judgment, planning, and exceptions.
When to use this skill
When a task needs the physical job "patrol water and wastewater plants, read gauges, take samples, turn valves, and service pumps and dosing systems" in environments such as water/wastewater treatment plants, pump stations, chemical dosing rooms. Pair with the sector skill (../../SKILL.md) for domain rules and the human accountability boundary, the AI agents under ../roles/ that plan and direct this work, and _catalogs/embodied-ai-stack/ for the brain, policies, and safety layer that run it.
Cognitive and control architecture (assumed)
These robot roles are assumed to be LLM-brained embodied agents, not hard-coded automatons. The stack:
- Cognitive core (the "brain"). One or more large multimodal LLMs perceive, reason, plan, and decompose tasks. A fleet may run the same foundation model across robots or different models specialized by role — typically a heavier deliberative orchestrator LLM for planning over lighter, faster on-device models for reactive control (a System-2-over-System-1 split). The brain is interchangeable and upgradable independent of the body.
- Actions are tool calls. Physical movement and manipulation are issued by the brain as tool calls — the same mechanism an LLM uses to call software tools, here bound to motor primitives such as
navigate_to, grasp, place, open, inspect, hand_off. The brain decides what; lower-level policies execute how.
- Low-level control: Vision-Language-Action (VLA) policies. Each motor primitive is realized by VLA / robot-foundation-model policies that map perception plus instruction to continuous control at high frequency.
- Trained on world models + robot gyms. Planners and policies are trained against world models (learned predictive simulators of physics and outcomes, used to imagine consequences before acting) and robot gyms (massively parallel physics simulation for sim-to-real skill learning), then transferred to hardware.
- RLAIF (RL from AI Feedback) — one method among many. Skills can be refined with reinforcement learning where an AI critic supplies reward and preference signals at scale, but RLAIF is only one option: imitation/behavior cloning, model-based and offline RL, sim-to-real, supervised fine-tuning, and distillation/compression into SLMs and tiny LMs all contribute, with deterministic controllers for hard-real-time, safety-critical loops. The brain is right-sized per task — LLM ↔ SLM ↔ tiny LM ↔ deterministic. See
_catalogs/capability-optimization/.
Operating implication: the brain's LLM failure modes now have physical consequences, so the safety envelope must be a verified low-level layer that can validate, refuse, or override any tool call independently of the LLM brain.
Division of labor and safety
- Human owner (certified plant operator / utility manager) — owns water quality and public-health compliance, chemical dosing decisions, safety, and exceptions; holds override and stop authority.
- LLM brain — perceives the plant/site, plans the task, and issues motor-primitive tool calls (
navigate_to, grasp, pick, place, inspect).
- VLA policies — execute dexterous, delicate manipulation (e.g., capturing an uncontaminated grab sample) under the engineered safety envelope.
- AI agents — the sector's planning/monitoring agents (treatment optimization, water-quality monitoring, asset-maintenance planning) direct and schedule the robot's work.
- Verified safety layer — validates, refuses, or overrides unsafe tool calls independently of the brain (people and the water supply protected).
Accountability boundary
Public health notices, water shutoffs, infrastructure investment, environmental-discharge approvals, and emergency allocation remain human-led.
These remain human-owned. The robot executes within an engineered envelope and routes anything outside it — water-quality anomalies, chemical-handling decisions, or unsafe conditions — to the accountable human.
Operating and safety procedure
- Confirm the plant/site is mapped, people and the water supply are protected, and the task is within the engineered envelope.
- The brain plans and emits motor-primitive tool calls; the safety layer validates each before execution.
- Execute within speed, force, reach, and contamination-control limits via VLA policies.
- Report progress, outcomes, exceptions, and any safety event to the sector agents and human owner.
- Stop and yield to humans for out-of-distribution conditions, contamination risk, or anything outside the envelope.
Architecture-specific failure modes
- Hallucinated or unsafe tool calls — the LLM brain issues a wrong or dangerous action. Mitigation: a verified low-level safety layer that validates every tool call against the physical envelope and can refuse it.
- Sim-to-real gap — world-model / robot-gym training diverges from reality. Mitigation: conservative behavior on out-of-distribution inputs, real-world evaluation, graceful degradation.
- Reward hacking from RLAIF — the AI critic is gamed, yielding behavior that scores well but is unsafe. Mitigation: diverse critics, human spot-checks, outcome-based evaluation.
- Physical-world prompt injection — adversarial signs, audio, or objects manipulate the brain. Mitigation: treat perceived instructions as untrusted; require authenticated commands for high-consequence actions.
- Fleet model-monoculture — a shared brain fails in lockstep across many robots. Mitigation: model diversity, staged rollouts, manual fallback.
Labor-market grounding (how these roles are advertised)
The human roles this operating system staffs appear on job boards with concrete, checkable signals. The AI-personnel and robot skills here are designed to support these advertised roles, not to replace the accountable human in them.
- Advertised titles & seniority ladder: Operator trainee → certified operator (Grade I–IV) → chief operator/superintendent → utility director; engineering: EIT → PE.
- Skills, tools & tech employers list: SCADA, GIS, hydraulic modeling (EPANET, WaterGEMS), LIMS, CMMS (asset/maintenance), telemetry.
- Qualifications, certifications & licenses: State water/wastewater operator certification (Grades I–IV), PE (civil/environmental), backflow tester, confined-space, CDL (some).
- KPIs / metrics in postings: Water-quality compliance, non-revenue water/leakage, NPDES permit compliance, boil-water/outage events, asset condition.
- Where these roles are posted: GovernmentJobs, Careers..gov, AWWA/WEF job boards, Indeed, ZipRecruiter.
Grounding reflects 2026 job-posting conventions across LinkedIn, Indeed, Dice, ZipRecruiter, Glassdoor, USAJOBS, GovernmentJobs, and specialized boards, spot-verified against public listings and O*NET/BLS. Re-verify specifics — especially pay, certifications, and licenses — against live postings before operational use.
Adapting to any nation (context modifiers)
In low-capacity settings this role may serve many small plants on a circuit, or support community-managed water points where the "plant" is a borehole and a tank; robustness and simple field maintenance dominate. In high-income utilities it extends around-the-clock coverage of aging assets and keeps humans out of confined and chemical spaces. Re-read through:
- Scale (city-state → federation): whether this role is unified or layered across local/regional/national tiers.
- State capacity (fragile → high-capacity): whether the owning institution exists and can be held to account, or the job is met by markets, households, NGOs, or donors.
- Income level (low → high): affordability of automation and the balance of subsistence vs. wage work.
- Formality (informal → formal): whether the people and assets this role acts on appear in any registry at all.
- Resource & geography: which hazards and dependencies dominate (water-scarce, flood-prone, landlocked, trade-dependent).
- Political system & legitimacy: where the human-accountability boundary actually binds and who may hold power to account.
