Role Definition
| Field | Value |
|---|---|
| Job Title | Clean Water Process Engineer |
| SOC Code | 17-2081 (Environmental Engineers) / 17-2041 (Chemical Engineers) |
| Seniority Level | Mid-Level (3-8 years, PE/CEng obtained or in progress) |
| Primary Function | Designs drinking water treatment processes — filtration (rapid gravity, slow sand, membrane), UV disinfection, chlorination/chloramination, ozone, activated carbon, and advanced membrane processes (RO, NF, UF). Performs hydraulic and process modelling using EPANET, WaterGEMS, or InfoWorks WS. Analyses raw and treated water quality data for regulatory compliance monitoring. Conducts site investigations and pilot testing for treatment process selection. Writes technical reports, design documentation, and basis-of-design documents. Liaises with regulators on DWI (UK) or EPA Safe Drinking Water Act (US) compliance. Provides construction oversight and commissioning support for treatment works upgrades. Splits time primarily between office-based modelling/design and periodic site visits for pilot testing and commissioning. Works for water utilities (Scottish Water, Thames Water, Southern Water) or consultancies (Tetra Tech, Stantec, Mott MacDonald, AECOM). |
| What This Role Is NOT | NOT a Wastewater Process Engineer (who designs activated sludge, MBR, and nutrient removal systems for effluent treatment — scored 50.1 Green). NOT a Water/Wastewater Treatment Plant Operator (who runs daily plant operations and adjusts chemical dosing — scored 52.4 Green). NOT an Environmental Engineer generalist (broader scope including remediation, air quality, waste — scored 40.3 Yellow). NOT a Water Resources Engineer (who focuses on hydrology, flood management, and catchment-scale water supply). NOT a junior process engineer doing primarily data compilation and standard calculations under supervision. |
| Typical Experience | 3-8 years. ABET-accredited (US) or IChemE/ICE-accredited (UK) bachelor's or master's in environmental, chemical, or civil engineering with water treatment focus. FE exam passed / working toward Chartered status. PE license (US) or CEng (UK) obtained or in progress — required for design sign-off. Proficiency in EPANET, WaterGEMS, or InfoWorks WS. Working knowledge of filtration design, membrane processes, disinfection by-product management, and drinking water quality standards. |
Seniority note: Junior process engineers (0-2 years) doing primarily data entry into hydraulic models, standard sizing calculations, and report drafting under supervision would score deeper Yellow (low 40s). Senior/principal engineers with PE/CEng stamps, client relationships, DWI/EPA regulatory negotiation authority, and emerging contaminant treatment specialisation would score Green (low 50s).
Protective Principles + AI Growth Correlation
| Principle | Score (0-3) | Rationale |
|---|---|---|
| Embodied Physicality | 1 | Site investigations, pilot testing at treatment works, and commissioning support require physical presence — but less frequently than wastewater counterparts. Clean water process design is more desk-based: hydraulic modelling, data analysis, and report writing dominate daily work. Pilot testing of membrane or UV systems occurs in structured plant environments with known configurations. |
| Deep Interpersonal Connection | 1 | Coordinates with water utility clients, regulators (DWI, EA, EPA), construction teams, and operations staff. Communicates treatment recommendations to non-technical stakeholders. Important but primarily technical — trust built through competence and regulatory knowledge, not empathy. |
| Goal-Setting & Moral Judgment | 2 | Treatment process design directly affects public health — inadequate drinking water treatment means contaminated supply reaching homes, hospitals, schools. Interpreting ambiguous raw water quality data (seasonal turbidity spikes, emerging contaminant detections), balancing treatment performance against capital cost, and making professional judgment calls about design safety factors require experienced engineering judgment. DWI prosecutions (UK) and EPA enforcement actions (US) create personal professional accountability. The Flint water crisis demonstrated catastrophic consequences of water treatment engineering failures. |
| Protective Total | 4/9 | |
| AI Growth Correlation | 0 | Demand driven by aging water infrastructure, population growth, tightening drinking water standards (PFAS MCLs, Lead and Copper Rule revisions), and climate-related raw water quality changes — not AI adoption. AI tools augment modelling and data analysis but do not proportionally create or eliminate positions. Neutral. |
Quick screen result: Protective 4/9 with neutral growth and strong regulatory accountability — likely borderline Yellow/Green. Proceed to quantify.
Task Decomposition (Agentic AI Scoring)
| Task | Time % | Score (1-5) | Weighted | Aug/Disp | Rationale |
|---|---|---|---|---|---|
| Treatment process design (filtration, UV, membrane, chemical) | 25% | 2 | 0.50 | AUGMENTATION | Designing drinking water treatment trains — selecting filtration type (rapid gravity, slow sand, membrane), disinfection strategy (UV, chlorination, ozone), membrane configuration, and chemical dosing systems. Requires integrating raw water quality characteristics, seasonal variability, DWI/EPA standards, site constraints, and lifecycle cost. AI can generate preliminary sizing and explore design alternatives, but technology selection for site-specific raw water conditions, design safety factor decisions, and constructability judgment require experienced engineering judgment. Less biological complexity than wastewater (no activated sludge kinetics) but significant physicochemical complexity in membrane fouling, DBP formation potential, and emerging contaminant treatment. |
| Hydraulic/process modelling (EPANET, WaterGEMS) | 15% | 3 | 0.45 | AUGMENTATION | Network hydraulic modelling and treatment process simulation. AI-enhanced optimisation of network pressures, pump scheduling, and treatment process parameters is maturing. EPANET's deterministic hydraulic calculations and WaterGEMS scenario analysis are well-suited to AI acceleration. Engineer still required for model setup, demand allocation, calibration against field measurements, and interpreting results for design decisions — but routine model runs and sensitivity analyses are increasingly automated. More automatable than BioWin/GPS-X wastewater modelling because hydraulic models are more deterministic than biological process models. |
| Water quality data analysis/compliance monitoring | 15% | 3 | 0.45 | AUGMENTATION | Analysing raw and treated water quality datasets, monitoring regulatory compliance against DWI/EPA standards, tracking disinfection by-product trends, identifying seasonal patterns in turbidity, colour, and microbiology. AI excels at pattern recognition in time-series water quality data, anomaly detection, and automated compliance checking. Engineer required for interpreting anomalies, determining root causes, and recommending treatment adjustments — but the analytical heavy lifting is increasingly AI-assisted. |
| Site investigations/pilot testing | 15% | 2 | 0.30 | NOT INVOLVED | Conducting site visits to assess existing treatment works, evaluating raw water sources, and running pilot-scale treatment tests (membrane pilot rigs, jar testing, UV dose-response trials). Requires physical presence at treatment works. Clean water pilot testing is less intensive than wastewater — treatment works are cleaner, more structured environments; pilot durations are typically shorter; and membrane/UV pilot testing is more standardised than biological process pilots. No AI involvement in the physical execution. |
| Technical report writing/design documentation | 15% | 4 | 0.60 | DISPLACEMENT | Engineering reports, basis-of-design documents, technical memoranda, options appraisal reports, O&M manuals. AI generates substantial portions from process data, design parameters, and templates. Standard documentation is highly automatable — particularly options appraisals and O&M manuals. Engineer reviews, certifies, and adds engineering judgment commentary but does not draft from scratch. |
| Regulatory liaison/DWI/EPA compliance | 10% | 2 | 0.20 | AUGMENTATION | Engaging with DWI (UK) or EPA (US) on treatment requirements, consent conditions, and compliance strategies. Interpreting Safe Drinking Water Act MCLs, Water Supply (Water Quality) Regulations, and guidance for novel scenarios (PFAS, lead, manganese, Cryptosporidium). AI assists with regulatory database searches and compliance tracking, but interpreting standards for site-specific conditions, negotiating with regulators, and certifying compliance require PE/CEng-level professional judgment. |
| Construction oversight/commissioning support | 5% | 1 | 0.05 | NOT INVOLVED | Witnessing hydraulic testing of filter beds, membrane installations, and UV systems. Providing technical support during construction and commissioning of treatment works upgrades. Coordinating with contractors on-site during critical commissioning stages. Physically present at treatment works for pressure testing, chemical cleaning, and system start-up. No AI involvement in physical oversight and commissioning — this is hands-on engineering presence. |
| Total | 100% | 2.55 |
Task Resistance Score: 6.00 - 2.55 = 3.45/5.0
Displacement/Augmentation split: 15% displacement, 65% augmentation, 20% not involved.
Reinstatement check (Acemoglu): Moderate reinstatement. AI creates new tasks: validating AI-optimised chemical dosing recommendations against raw water variability, interpreting ML-driven anomaly detection in water quality datasets, designing treatment systems for novel contaminants (PFAS, microplastics, pharmaceuticals) where AI training data is sparse, auditing AI-generated hydraulic model calibrations, and managing digital twin systems for treatment works optimisation. The role shifts from manual model runs and report drafting toward judgment-intensive validation, emerging contaminant engineering, and AI-augmented design optimisation.
Evidence Score
| Dimension | Score (-2 to 2) | Evidence |
|---|---|---|
| Job Posting Trends | 1 | BLS projects 4% growth 2024-2034 for environmental engineers (17-2081). Drinking water treatment sub-specialty growing faster driven by EPA PFAS MCL compliance deadlines, Lead and Copper Rule revisions, IIJA infrastructure funding ($50B+ for water), and aging treatment works requiring upgrades. UK demand strong from AMP8 investment cycle (2025-2030). AWWA reports persistent workforce shortages in water sector engineering. |
| Company Actions | 1 | No engineering firms cutting clean water process engineers citing AI. Major water engineering consultancies (Stantec, Tetra Tech, Mott MacDonald, AECOM, Jacobs, Binnies) continue active hiring for drinking water process roles. Scottish Water, Thames Water, United Utilities investing in treatment works upgrades. PFAS treatment projects expanding project pipelines. |
| Wage Trends | 1 | UK: £35,000-£55,000 mid-level; US: $65,000-$90,000 mid-level. BLS environmental engineer median $104,170 (May 2024). Growing above inflation. PE/CEng-licensed process engineers in drinking water command premiums, particularly for PFAS and advanced membrane treatment specialisations. Solid wage growth driven by infrastructure investment and moderate talent shortage. |
| AI Tool Maturity | 0 | EPANET and WaterGEMS adding AI-enhanced optimisation features. AI-optimised chemical dosing systems emerging in pilot (real-time coagulant dosing based on turbidity prediction). Digital twins for water treatment works in early adoption. But process design and technology selection remain human-led. No commercial AI tools performing autonomous drinking water treatment design. Tools augment modelling workflow without replacing design judgment. |
| Expert Consensus | 0 | Mixed but leaning positive. AWWA and ICE describe AI as tool for treatment optimisation, not engineer replacement. ASCE reports slow AI adoption in engineering (27% of firms). No credible source predicts clean water process engineer displacement. However, no strong consensus on demand acceleration beyond infrastructure investment cycle and regulatory mandates. |
| Total | 3 |
Barrier Assessment
Reframed question: What prevents AI execution even when programmatically possible?
| Barrier | Score (0-2) | Rationale |
|---|---|---|
| Regulatory/Licensing | 2 | PE license (US) or CEng (UK) required for stamping drinking water treatment designs. DWI (UK) requires competent person sign-off on treatment process changes. EPA Safe Drinking Water Act requires PE-certified professionals for public water supply design. State boards mandate PE for design of public drinking water infrastructure. Among the strongest licensing barriers in engineering — you cannot legally design a treatment works without PE/CEng authorisation. |
| Physical Presence | 1 | Pilot testing, site investigations, and commissioning require physical presence at treatment works. Site visits to observe process conditions (filter media, membrane integrity, UV lamp condition) are routine. But majority of daily work is office-based modelling and design. Less physically embedded than treatment plant operators. Less site time than wastewater process engineers who spend more time on biological pilot testing. |
| Union/Collective Bargaining | 0 | Clean water process engineers in consulting and utilities are not typically unionised. No collective bargaining protection. |
| Liability/Accountability | 2 | PE/CEng-stamped treatment designs carry personal professional liability. Inadequate drinking water treatment causes direct public health emergencies — contaminated supply reaching homes, hospitals, schools. DWI prosecutions (UK), EPA enforcement actions (US), and potential criminal liability create strong personal accountability. The Flint water crisis, Walkerton E. coli outbreak, and Cryptosporidium incidents demonstrated that drinking water engineering failures carry severe legal consequences. Drinking water liability is arguably stronger than wastewater because contamination reaches consumers directly. |
| Cultural/Ethical | 1 | Public expects human engineers designing drinking water treatment systems that protect community health. Regulators expect PE/CEng-certified professionals certifying treatment designs. Moderate cultural resistance to AI designing critical public health infrastructure. Drinking water carries emotional weight — "what comes out of my tap" — creating public sensitivity to non-human design authority. |
| Total | 6/10 |
AI Growth Correlation Check
Confirmed at 0 (Neutral). Drinking water treatment infrastructure demand is driven by aging water systems (UK AMP8 investment, US IIJA $50B+), tightening drinking water standards (EPA PFAS MCLs, Lead and Copper Rule revisions), population growth, and climate-related raw water quality changes — not AI adoption. AI tools make existing process engineers more productive at modelling and data analysis, but the demand signal is infrastructure and regulatory, not technological. Neither accelerated nor diminished by AI growth.
JobZone Composite Score (AIJRI)
| Input | Value |
|---|---|
| Task Resistance Score | 3.45/5.0 |
| Evidence Modifier | 1.0 + (3 x 0.04) = 1.12 |
| Barrier Modifier | 1.0 + (6 x 0.02) = 1.12 |
| Growth Modifier | 1.0 + (0 x 0.05) = 1.00 |
Raw: 3.45 x 1.12 x 1.12 x 1.00 = 4.3277
JobZone Score: (4.3277 - 0.54) / 7.93 x 100 = 47.8/100
Zone: YELLOW (Green >=48, Yellow 25-47, Red <25)
Sub-Label Determination
| Metric | Value |
|---|---|
| % of task time scoring 3+ | 45% (modelling 15% + data analysis 15% + reporting 15%) |
| AI Growth Correlation | 0 |
| Sub-label | Yellow (Urgent) — 45% >= 40% threshold |
Assessor override: None — formula score accepted. At 47.8, this sits 0.2 points below the Green threshold and 2.3 points below the Wastewater Process Engineer (50.1). The gap is justified: drinking water treatment chemistry is more deterministic than wastewater biological processes (activated sludge kinetics, microbial community variability), meaning AI process modelling is more viable for standard drinking water configurations. The 0.15 TR difference (3.45 vs 3.60) reflects less pilot testing complexity (membrane and UV systems are more standardised than biological treatment) and more desk-based daily workflow. The barriers (6/10) are identical to the wastewater peer — PE licensing and public health liability are equally strong for both. Compare to Environmental Engineer (40.3 Yellow) — the 7.5-point gap reflects stronger barriers (6/10 vs 4/10, PE more universally required in water treatment design), stronger evidence (+3 vs +2), and higher task resistance (3.45 vs 3.20).
Assessor Commentary
Score vs Reality Check
The Yellow (Urgent) classification at 47.8 is honest but borderline. This role sits 0.2 points below Green, making it the closest Yellow-to-Green boundary case in the water sector. Task resistance (3.45) is lower than the wastewater counterpart (3.60) because clean water process design involves less biological complexity and more deterministic modelling: EPANET hydraulic calculations are inherently more automatable than BioWin activated sludge kinetics. The barriers (6/10) are doing significant work — PE/CEng licensing is near-universal for drinking water treatment design, and the liability for public health contamination through the drinking water supply is among the most severe in engineering. Without the barrier boost, the score would be 42.5 (mid-Yellow) — the PE and liability barriers are what bring this to the Green threshold. The evidence (+3) reflects genuine demand tailwinds from PFAS MCLs, AMP8 investment (UK), IIJA funding (US), and Lead and Copper Rule revisions.
What the Numbers Don't Capture
- PFAS as demand accelerator — EPA's final PFAS MCLs (4 ppt for PFOA/PFOS) are forcing thousands of US water systems to install new treatment technologies. PFAS treatment design (GAC, ion exchange, high-pressure membranes, PFAS destruction technologies) is a rapidly growing sub-speciality where AI training data is minimal. Engineers specialising in PFAS drinking water treatment face demand not yet reflected in BLS projections.
- Deterministic modelling as vulnerability — Unlike wastewater's biological complexity moat, clean water hydraulic and physicochemical modelling is more deterministic, making it a better target for AI automation. Engineers who rely primarily on running EPANET models and standard design calculations face greater automation pressure than those who combine modelling with field-based pilot testing and emerging contaminant expertise.
- Climate-driven raw water quality changes — Increasing turbidity events, algal blooms (taste and odour), and seasonal temperature extremes are making treatment process design more complex and variable. Engineers who understand how to design resilient treatment trains for variable raw water quality are increasingly valued.
- Function-spending vs people-spending — AI-augmented modelling may enable individual engineers to handle more projects, potentially limiting headcount growth even as project volumes increase. This is more pronounced in clean water than wastewater because the modelling tools are more mature and deterministic.
Who Should Worry (and Who Shouldn't)
Clean water process engineers who hold PE/CEng licences, conduct pilot testing for membrane and advanced treatment systems, specialise in emerging contaminants (PFAS, lead, manganese, Cryptosporidium), and provide commissioning support on-site are well-protected — their value comes from physical-world engineering judgment, professional accountability, and specialised knowledge in areas where AI tools are least mature. Process engineers whose daily work is primarily running standard EPANET/WaterGEMS models for conventional treatment works, producing options appraisal reports, and performing standard filter or UV sizing calculations without PE/CEng stamps or field involvement are more exposed — AI-enhanced modelling tools directly target these workflows. The single biggest differentiator is PE/CEng licensure combined with emerging contaminant expertise: a chartered engineer who has designed PFAS treatment systems and commissioned membrane plants is deeply protected. A non-chartered engineer doing standard hydraulic modelling and report writing at a desk is vulnerable to productivity compression. Engineers who pivot into PFAS treatment design, advanced membrane processes, and emerging regulatory compliance have the strongest demand trajectory.
What This Means
The role in 2028: Mid-level clean water process engineers spend less time on routine hydraulic model runs, standard treatment sizing calculations, and boilerplate report sections as AI-enhanced modelling and documentation tools mature. More time shifts to designing treatment systems for novel contaminants (PFAS, microplastics, pharmaceuticals), validating AI-optimised chemical dosing systems, leading pilot studies for advanced membrane and oxidation technologies, and interpreting AI-driven water quality analytics. The engineer who masters AI-augmented modelling tools becomes more productive — running more design scenarios faster and providing better-informed treatment technology recommendations.
Survival strategy:
- Obtain your PE/CEng licence. The professional stamp is the single strongest differentiator between protected and exposed clean water process engineers. It creates personal liability, regulatory authority, and an institutional barrier AI cannot cross. If you are working toward PE/CEng, prioritise completing it.
- Specialise in PFAS and advanced treatment. PFAS MCL compliance, advanced membrane design (NF/RO), and advanced oxidation processes are where demand is growing fastest and AI tools are least mature. Avoid becoming a generalist running standard coagulation-filtration models.
- Build pilot testing and commissioning experience. Physical field work — operating membrane pilot rigs, commissioning UV systems, conducting jar testing at treatment works — is the AI-resistant core of this role. Seek projects that put you on-site, not just behind a hydraulic model.
Where to look next. If you're considering adjacent roles, these share transferable skills:
- Wastewater Process Engineer (Mid-Level) (AIJRI 50.1) — Biological treatment systems add complexity that strengthens AI resistance. Process modelling, PE licensing, and design judgment transfer directly.
- Water and Wastewater Treatment Plant Operator (Mid-Level) (AIJRI 52.4) — For process engineers wanting maximum physical-world protection. Process engineering knowledge is highly valued in operations.
- Environmental Engineer (Mid-Level) (AIJRI 40.3) — Broader environmental scope but currently Yellow. Weaker PE mandates and less direct infrastructure investment tailwind.
Browse all scored roles at jobzonerisk.com to find the right fit for your skills and interests.
Timeline: 3-7 years for significant transformation of modelling and reporting workflows. Pilot testing, commissioning, and PE/CEng-stamped design work persist indefinitely. PFAS compliance deadlines (2029 US) and infrastructure investment cycles (UK AMP8, US IIJA $50B+) provide a sustained demand floor through the mid-2030s.
