Technical report: MTF-CF-RR-2026-10-01-57
Institution: MTF Institute
Evidence date: 30 September 2026
Publication date: 1 October 2026
Geography: United States

Executive summary

This report examines what employers asked occupational health and safety coordinators and closely comparable safety specialists to do in 100 current U.S. vacancies retrieved on 30 September 2026. The vacancies came from 81 named employer identities. The underlying ledger contains 78 literal employer labels; three labels name two organizations in a joint or group relationship, so the employer-identity count is 81. Every accepted vacancy has a unique detail URL and a documented requisition identity or provenance.

The study is a structured purposive, point-in-time analysis. It is not a statistically representative survey of all U.S. employers, a measure of the number of available jobs, a hiring forecast, or a prediction of an applicant’s prospects. Public English-language and ATS-visible vacancies are easier to observe than other recruiting channels. Source concentration also matters: 39 accepted records were hosted on the exact domain www.career.com, and 22 were Amazon requisitions. Those concentrations are disclosed and tested rather than hidden.

The most frequently stated work pattern was coordination with supervisors, operations leaders or project leadership, found in 85 of 100 vacancies. Records, metrics or reporting appeared in 68; workplace observations, inspections or assessments in 63; and incident or near-miss intake or investigation in 61. Hazard reporting and intake appeared in 51, worker participation in 50, corrective-action follow-through in 39, and toolbox talks or safety meetings in 25. Escalation or professional-boundary routing appeared in 53. Explicit blame-aware or learning-team language appeared in only two vacancies. That low count should not be mistaken for evidence that careful, systems-oriented incident learning is unimportant; it means only that this language was rarely explicit in the retained vacancy text.

The combined findings describe an operating role built around a recurring loop: observe and receive information, preserve it in usable records, communicate with workers and supervisors, track agreed actions, and route decisions beyond the coordinator’s authority. The sample does not support treating an ordinary coordinator as a universal legal, engineering, medical, industrial-hygiene or emergency authority. Employer language is evidence of recruitment expectations, not professional advice. Applicable federal or OSHA-approved State Plan requirements, local rules, employer procedures, collective-bargaining terms, site conditions and decisions by qualified professionals continue to govern actual work.

Four findings retained the same frequency band across all source-sensitivity subsets: supervisor routines, records and metrics, corrective-action closure, and explicit blame-aware incident learning. Six other focus areas changed bands in at least one subset. Their full-sample counts remain valid descriptions of this accepted corpus, but they should not be generalized as national prevalence.

Research question and role boundary

The research question was: What duties, outputs, skills, tools, qualifications, working rhythms, interfaces, decisions and escalation boundaries are stated in current U.S. vacancies for occupational health and safety coordination work?

The target population included public U.S.-located vacancies for Safety Coordinator, EHS Coordinator, Occupational Health and Safety Coordinator and narrowly comparable specialist roles when their work remained coordination-equivalent. A role needed enough public information to code its duties, outputs, authority and escalation. Senior managers, directors, industrial hygienists, engineers and specialist roles were excluded when their independent technical authority materially exceeded the intended boundary. Roles outside the United States, expired pages, unrelated safety meanings and records without defensible currentness were also excluded.

The retained corpus contains 26 entry or developmental roles, 68 general coordinator or specialist roles, and six experienced or edge-of-scope roles. Five records received an explicit include-with-limit decision because part of the title, experience requirement or responsibility sat at the upper edge of the population. Those records were kept as progression evidence without allowing their exceptional responsibilities to define the early-career core. A sixth role was coded at the experienced level but remained within scope without a special limitation.

This boundary is important because occupational health and safety titles do not create identical authority across employers. A coordinator may collect facts, identify conditions, support a briefing, maintain a log or follow up an action. The accountable decision about a technical control, medical response, legal notification, exposure assessment, equipment state, recordability, work release or emergency response may belong elsewhere.

Corpus construction

Accepted records and deduplication

The final corpus contains exactly 100 accepted current U.S. vacancies. All 100 have unique canonical detail URLs. Deduplication used employer, source-issued or publisher identity, normalized title and location, and the canonical detail URL. Two rediscovered records were logged as duplicates rather than counted again: a Lever vacancy found through another search path and an Amazon category page used only for discovery. Sixteen other records were excluded for reasons including expired status, non-U.S. location, an unrelated meaning of safety, scope above the target role, or inadequate evidence on the cited page.

Currentness was supported by an active application or detail page together with a dated posting, recent update or explicit availability signal at retrieval. A page becoming unavailable later would not retroactively change what was observed on the evidence date, but it would limit later reproducibility.

Employer counting

The accepted records use 78 distinct literal employer labels. Three labels each identify two named organizations in a joint, subsidiary or group relationship: Network Connex / NTI, TAS Energy / Comfort Systems USA, and Wanzek Construction / MasTec Renewables. Counting the separately named organizations in those labels produces 81 employer identities. Amazon accounts for 22 records and Northrop Grumman for two; no other literal employer label contributes more than one vacancy.

Source-family composition

The sample combines employer-controlled sources with public secondary detail pages. Secondary pages were not treated as interchangeable with employer pages; coding was limited to facts retained from the specific accepted source.

Accepted source family Vacancies
Job aggregator mirroring employer vacancy 33
First-party employer careers 24
First-party employer ATS 14
Staffing vacancy 6
Professional EHS job board detail page 4
Staffing / job aggregator 4
First-party employer ATS listing 2
Job aggregator / staffing 2
Higher-education job board 1
Industry job board 1
Industry job board mirroring employer vacancy 1
Job aggregator 1
Job board 1
Job mirror with live-source check 1
Job mirror with live-source signal 1
Professional association job board 1
Professional EHS job board / staffing detail page 1
Public-sector vacancy mirror 1
Staffing / project vacancy 1
Total 100

Forty vacancies form the deterministic direct or employer-authorized subset: 14 first-party employer ATS records, 24 first-party employer-careers records and two first-party employer ATS listings. No host or publisher guess was used to enlarge that subset.

Sector concentration

The work-situation coding places 54 vacancies in manufacturing or general industry, 22 in warehouse or logistics settings, 18 in construction or project work, four in industrial field work, and two in laboratory or technical settings. These mutually exclusive coding buckets show that the study is weighted toward operational environments, especially manufacturing and logistics. They do not describe the sector distribution of the whole U.S. occupation.

Coding and interpretation rules

The unit of analysis was one unique vacancy. Each record was coded across duties, outputs, hard skills, behavioral skills, tools, education and experience, required criteria, preferred criteria, level, cadence, interfaces, authority, escalation, decisions, work situations and ten focus areas. A positive focus code had to point to a source-specific supporting fact. Missing information was coded as not_stated; silence was never converted into a negative finding or filled from a generic role profile.

Frequency bands were applied consistently: high frequency means at least 60% of the accepted sample, medium frequency means 20% to 59.9%, and low frequency means below 20%. “High criticality” identifies a safety-significant requirement retained for interpretation regardless of frequency. “Emerging” identifies explicit newer workflow or vocabulary and is not a growth prediction.

Counts within dimensions may overlap. One vacancy can mention inspection, incident intake, records and supervisor coordination at the same time. Experience buckets also overlap by design when a stated range crosses the threshold. Percentages therefore should not be added to make a total.

What the vacancies say about the work

Focus-area findings

Focus area Count Share Band Safety-significant flag
Supervisor routines 85 85% High No
Records and metrics 68 68% High No
Workplace observation and inspection 63 63% High No
Incident and near-miss intake 61 61% High Yes
Escalation and professional-boundary routing 53 53% Medium Yes
Hazard reporting and intake 51 51% Medium Yes
Worker participation 50 50% Medium No
Corrective-action ownership and closure 39 39% Medium Yes
Toolbox talk / safety meeting preparation and facilitation 25 25% Medium No
Blame-aware incident learning 2 2% Low Yes

Hazard reporting is an intake-and-routing responsibility

Hazard reporting and intake appeared in 51 vacancies. This finding is best read together with the 53 vacancies that stated escalation or professional routing and the 50 that included worker participation. A coordinator’s contribution is not merely to provide a form. It is to make a concern receivable, preserve enough factual information for the next decision, acknowledge or route it, maintain status visibility and help close the feedback loop.

The vacancies do not establish one universal reporting channel. Employers may use supervisor conversations, observations, digital systems, committees, hotline routes, incident systems or local forms. The transferable requirement is disciplined intake and follow-through while protecting confidentiality and avoiding reporting suppression. Examples and metrics should not blame a worker for raising a concern or create pressure to keep reported numbers artificially low.

Observation and inspection connect the field to the record

Workplace observation, inspection, audit or assessment appeared in 63 vacancies. The same count appears for hazard-recognition and inspection methods, inspection forms, inspection outputs and decisions about observed conditions. This alignment shows a coherent chain in the coding: the duty is performed with a method or form, it produces a record, and it supports identification of a condition requiring action.

That chain does not confer unlimited technical authority. A coordinator can document what was observed, compare it with the authorized site method, identify the owner of the next step and escalate uncertainty. Whether a control is technically adequate, an exposure is acceptable, equipment may return to service or specialized testing is sufficient can require an engineer, industrial hygienist, competent person, qualified person, medical professional or other accountable owner.

Toolbox talks and safety meetings are important but not universal

Twenty-five vacancies expressly assigned preparation or facilitation of toolbox talks, orientations, safety meetings or comparable training communication. The count places this work in the medium-frequency band in the full sample, but it falls to the low-frequency band in the direct/employer-authorized and excluding-Career.com subsets. The safest interpretation is that short safety communication is a meaningful part of some coordinator jobs, particularly in field and project settings, but it is not explicit across the whole corpus.

Vacancy evidence supports coordination and facilitation, not the unsupervised creation of site instructions for hazardous work. A briefing aid should be validated against the actual task, hazard, audience, language needs, controls, personal protective equipment, emergency arrangements and any demonstration required by the responsible employer or qualified site owner. A generic talk cannot replace training required for a particular standard, employer or site.

Incident and near-miss work emphasizes intake, evidence and handoff

Incident or near-miss intake and investigation appeared in 61 vacancies. Sixty-one vacancies also mapped to the related hard skill and incident-file output. The role therefore commonly touches the first usable record of an event: what happened, when and where it was observed, who holds relevant information, what evidence exists, what remains uncertain, and who must receive the handoff.

Only two vacancies used explicit blame-aware, learning-team or human-and-organizational-performance language. That narrow signal should be described exactly as observed. It does not support a claim that a new method is becoming common. It does, however, highlight a useful distinction between the prevalence of incident duties and the rarity of explicit systems-learning vocabulary in vacancy text.

Employer language is not a case-specific instruction. A coordinator should not diagnose an injury or illness, determine legal fault, decide regulatory recordability, make a workers-compensation decision, release someone to work or distribute sensitive medical or witness information. Immediate life-safety response follows the employer emergency plan and local emergency services. Later learning should preserve facts and uncertainty and route decisions to the accountable owner.

Corrective-action work means ownership visibility and closure evidence

Corrective-action follow-through or verification appeared in 39 vacancies. The matrix repeats that count across the duty, hard skill, decision and output dimensions. This consistency supports a bounded interpretation: coordinators are often expected to make action ownership visible, monitor due dates, seek evidence of completion, record status and raise overdue or ineffective responses.

The role is not automatically the technical designer or final approver of every control. “Closure” should mean that the agreed owner has completed the approved action and that the organization has retained suitable evidence or verification. It should not mean that a coordinator has independently certified compliance or overridden the authority of engineering, operations, occupational health, legal counsel or another specialist.

Supervisor routines form the strongest operating spine

Supervisor, operations or project coordination appeared in 85 vacancies, the highest count in the matrix. The same 85 records support cross-functional communication, interfaces with supervisors and decisions or activity organized around supervisor routines. This is also one of four focus areas whose frequency band remains stable in every source-sensitivity subset.

The evidence supports a practical rhythm rather than a generic people-management role. A coordinator may prepare an observation summary, surface open actions, confirm who owns a follow-up, support a briefing, check whether a record is complete and raise an unresolved issue. The supervisor retains line authority for staffing, work direction, performance and operational decisions unless the employer has assigned responsibilities differently. Safety coordination enables the routine and makes it reviewable; it does not replace supervision.

Records and metrics make work traceable

Records, logs, reports, metrics or dashboards appeared in 68 vacancies. Fifty-nine mentioned common office, spreadsheet or presentation tools; 29 mentioned a dedicated EHS or reporting platform. The evidence therefore supports tool adaptability rather than dependence on one named product. Inspection records, incident files, meeting records, action-status evidence and EHS logs or dashboards are recurring outputs.

Metrics should be interpreted within the vacancy evidence and the employer’s authorized system. The sample does not justify one universal dashboard or performance formula. A useful coordinator record distinguishes a fact from an inference, shows provenance and status, identifies an owner, protects restricted information and makes the next review possible. Counts that discourage injury, illness, hazard or close-call reporting are inconsistent with a healthy participation route.

Worker participation is operational, not decorative

Worker participation appeared in 50 vacancies, and coaching, listening or engagement also appeared in 50. This suggests that half of the accepted roles made a visible connection to frontline workers or safety committees. The source-sensitivity result changes across subsets, so the 50% figure should remain attached to this corpus rather than be described as national prevalence.

The practical implication is a two-way process: workers need an accessible way to contribute observations and concerns, and they need appropriate feedback about routing and follow-up. Communication is not only a broadcast. It also includes listening, clarifying, documenting uncertainty and returning status within privacy and authority limits.

Complete Role Requirements Matrix

The following tables report every matrix row. Repeated counts across dimensions show where the evidence linked a duty to a skill, decision, tool or output; they are not additional vacancies.

Duties, hard skills and behavioral skills

Dimension Requirement Count Share Band
Duty Inspections, audits, observations or assessments 63 63% High
Duty Training, orientation, toolbox talk or meeting facilitation 25 25% Medium
Duty Incident/near-miss intake or investigation 61 61% High
Duty Corrective-action follow-up or verification 39 39% Medium
Duty Records, metrics or reporting 68 68% High
Duty Supervisor/operations/project coordination 85 85% High
Hard skill Hazard recognition and inspection/audit methods 63 63% High
Hard skill Incident intake and investigation 61 61% High
Hard skill Corrective-action tracking and verification 39 39% Medium
Hard skill EHS recordkeeping and metrics 68 68% High
Hard skill Safety-meeting/training facilitation 25 25% Medium
Hard skill Learning-team/HOP methods 2 2% Low
Behavioral skill Cross-functional communication 85 85% High
Behavioral skill Coaching, listening and worker engagement 50 50% Medium
Behavioral skill Attention to detail and follow-through 68 68% High
Behavioral skill Problem solving and persistence 39 39% Medium
Behavioral skill Boundary-aware judgment 53 53% Medium

Tools, education, experience and selection criteria

Dimension Requirement Count Share Band
Tool Microsoft Office, spreadsheet or presentation tools 59 59% Medium
Tool Dedicated EHS or reporting platform 29 29% Medium
Tool Inspection, audit or risk-assessment forms 63 63% High
Education/experience Education requirement or preference stated 39 39% Medium
Education/experience Bachelor’s degree in required or accepted-pathway evidence 12 12% Low
Education/experience Degree expressly preferred 10 10% Low
Education/experience Entry experience up to two years stated 14 14% Low
Education/experience Three or more years stated 18 18% Low
Education/experience OSHA outreach or course credential stated 11 11% Low
Selection criteria One or more required criteria retained 21 21% Medium
Selection criteria One or more preferred criteria retained 19 19% Low

Education and experience were coded with explicit overlapping rules. The bachelor’s row includes a bachelor’s or B.S. degree stated as required or as a non-preferred accepted pathway; a degree mentioned only as preferred is excluded from that row. The degree-preferred row counts only explicit preference language. The zero-to-two-year row includes any required, preferred or alternative pathway whose minimum is zero, one or two years, including ranges that extend to three years. The three-plus row includes any pathway that reaches or begins at three years. A range can therefore contribute to both experience rows. Required and preferred criteria count only positive retained statements; silence remains not_stated.

The 11 references to an OSHA outreach or course credential record employer wording only. They do not make the role, this study or any education product OSHA-authorized. OSHA Outreach classes have their own provider and card rules and do not replace employer training required by specific standards.

Level, cadence, interfaces, authority and decisions

Dimension Requirement Count Share Band
Level Entry or developmental 26 26% Medium
Level General coordinator or specialist 68 68% High
Level Experienced or edge of scope 6 6% Low
Cadence Daily or shift-based 35 35% Medium
Cadence Weekly 11 11% Low
Cadence Monthly 8 8% Low
Cadence Event-driven 74 74% High
Cadence Off-shift or on-call 28 28% Medium
Interface Supervisors, operations or project leadership 85 85% High
Interface Frontline workers or safety committees 50 50% Medium
Interface Contractors or trade partners 18 18% Low
Interface EHS leadership, regulators, vendors or qualified specialists 53 53% Medium
Authority Authority or responsibility boundary stated 40 40% Medium
Escalation Escalation or professional routing stated 53 53% Medium
Decision Triage and route reported hazards or concerns 51 51% Medium
Decision Identify observed conditions requiring action 63 63% High
Decision Recommend follow-up and verify completion or effectiveness 39 39% Medium
Decision Distinguish coordinator action from escalation 53 53% Medium

The cadence data show both planned and reactive work. Event-driven activity was explicit in 74 vacancies, while 35 described daily or shift-based routines and 28 included off-shift or on-call expectations. Weekly and monthly rhythms were much less often stated. These categories overlap, so they should be used to design a balanced operating rhythm rather than to infer a standard schedule.

Only 40 vacancies stated an authority or responsibility boundary, while 53 described escalation or professional routing. Silence in the other records does not grant authority. The defensible cross-record interpretation is coordination, recommendation, tracking and escalation under employer-defined responsibility.

Work situations and outputs

Dimension Requirement Count Share Band
Work situation Manufacturing or general industry 54 54% Medium
Work situation Warehouse or logistics 22 22% Medium
Work situation Construction project 18 18% Low
Work situation Industrial field 4 4% Low
Work situation Laboratory or technical 2 2% Low
Output Inspection, audit or observation record 63 63% High
Output Training, orientation or meeting record 25 25% Medium
Output Incident or near-miss file 61 61% High
Output Corrective-action status or closure evidence 39 39% Medium
Output EHS log, metric, report or dashboard 68 68% High

These outputs show what makes coordination observable. The role is not defined only by attending meetings or walking a site. It produces evidence that another person can review: a clear observation record, an incident file with preserved facts, a meeting or orientation record, status and closure evidence for actions, and a log or report that supports the next decision.

Source-sensitivity tests

To test whether a concentrated source was driving the focus-area findings, the analysis recalculated all ten focus areas in four subsets: the full 100; the 40 direct or employer-authorized records; 61 records after excluding the exact host www.career.com; and 78 records after excluding Amazon. Exact parsed-host equality was used, so a different domain containing the word “career” was not removed.

Focus area Full Direct/employer-authorized Excluding www.career.com Excluding Amazon Band stable?
Hazard reporting and intake 51/100 (51%, medium) 15/40 (37.5%, medium) 24/61 (39.3%, medium) 47/78 (60.3%, high) No
Workplace observation and inspection 63/100 (63%, high) 34/40 (85%, high) 43/61 (70.5%, high) 43/78 (55.1%, medium) No
Toolbox talk / safety meeting 25/100 (25%, medium) 6/40 (15%, low) 11/61 (18%, low) 25/78 (32.1%, medium) No
Incident and near-miss intake 61/100 (61%, high) 28/40 (70%, high) 39/61 (63.9%, high) 40/78 (51.3%, medium) No
Blame-aware incident learning 2/100 (2%, low) 1/40 (2.5%, low) 1/61 (1.6%, low) 2/78 (2.6%, low) Yes
Corrective-action ownership and closure 39/100 (39%, medium) 9/40 (22.5%, medium) 19/61 (31.1%, medium) 38/78 (48.7%, medium) Yes
Supervisor routines 85/100 (85%, high) 33/40 (82.5%, high) 49/61 (80.3%, high) 66/78 (84.6%, high) Yes
Records and metrics 68/100 (68%, high) 33/40 (82.5%, high) 45/61 (73.8%, high) 47/78 (60.3%, high) Yes
Worker participation 50/100 (50%, medium) 30/40 (75%, high) 36/61 (59%, medium) 30/78 (38.5%, medium) No
Escalation and professional-boundary routing 53/100 (53%, medium) 14/40 (35%, medium) 29/61 (47.5%, medium) 51/78 (65.4%, high) No

The stable findings are especially useful for describing the corpus’s operating spine: supervisor coordination is consistently high, records and metrics remain high, corrective-action closure remains medium, and explicit blame-aware language remains low. The six unstable findings are still real counts within the full accepted sample. Instability means the assigned band depends on source composition, so those results need the subset table and cannot support a national prevalence claim.

The tests also show why a single percentage can mislead. Workplace observation rises to 85% in the direct subset but falls to 55.1% after Amazon is excluded. Hazard reporting moves in the opposite direction, reaching 60.3% without Amazon. Toolbox-talk language is 15% in direct sources and 32.1% without Amazon. Such differences can reflect sector, employer, job-family and source-description practices as well as the actual content of the work.

Professional, regulatory and rights boundaries

Employer vacancy language is recruitment evidence. It is not legal, medical, engineering, industrial-hygiene, emergency-response or compliance advice. Federal OSHA material provides U.S. federal context, but actual workplaces may be governed by federal OSHA or an OSHA-approved State Plan together with other federal, state and local rules, employer procedures, collective-bargaining terms and site-specific requirements.

This is a non-degree research publication. Reading it does not confer academic credit, licensure, certification, appointment or professional authority, including status as a competent person, qualified person, industrial hygienist, safety engineer, medical professional, incident commander or emergency responder. It does not issue an OSHA card or a BCSP, NEBOSH or IOSH credential. Employer references to such credentials are coded only as hiring language. Names of agencies, standards and professional bodies should be used only for accurate identification without suggesting approval, affiliation or equivalence.

Site-specific training remains an employer responsibility. A toolbox-talk draft, checklist or AI-assisted summary cannot by itself establish that a workplace is compliant or that a hazardous task may proceed. Hands-on decisions about equipment operation, exposure monitoring, confined spaces, energy control, fall protection, respiratory protection, electrical work, hazardous materials and emergency response require the applicable authorized procedure and qualified owner.

The research retained URLs, metadata, short necessary excerpts and independently written derived facts and codes. It did not retain full vacancy descriptions, applicant data, images or proprietary databases for redistribution. Employer and publisher expression should not be copied into educational or commercial materials, and no employer endorsement is implied.

Methodology and limitations

This was a structured purposive, point-in-time vacancy study. Public sources were collected without authentication or provider changes. The study prioritized current U.S. vacancies that supplied enough information to code the role. Each positive focus-area code was linked to a source-specific fact, and matrix counts were generated from record memberships. The silence policy treated an unstated item as missing information rather than as absence of a practice.

The method improves transparency but has important limits:

  • The sample is not statistically representative of all U.S. occupational health and safety vacancies, employers, sectors or regions.
  • Vacancy counts are not hires, employment stock, job availability, market share, a forecast, or evidence about any person’s employment prospects.
  • Public English-language and ATS-visible vacancies are more observable than closed, informal or inaccessible recruitment channels.
  • The sample is concentrated in manufacturing/general industry, warehouse/logistics and construction/project settings.
  • Source concentration is material: 39 records use the exact host www.career.com, and 22 are Amazon requisitions.
  • Secondary and mirrored pages vary in completeness. Coding was limited to retained source-specific facts.
  • Vacancy language may omit routine work, qualifications, tools or authority even when the employer expects them. not_stated is not zero.
  • Education, experience and selection-criteria counts depend on explicit wording and overlapping pathways. They do not define a universal entry standard.
  • Five explicit edge records were retained with scope limits, and six records were coded at the experienced or edge level. They show progression without defining the early-career core.
  • Only four of ten focus areas retained the same frequency band across all four sensitivity subsets. The other six require source-composition qualification.
  • Point-in-time vacancy pages can change or close after retrieval.
  • The study does not determine the legality, technical adequacy or safety of any workplace practice described by an employer.

Within those limits, the evidence supports a clear conclusion. Occupational health and safety coordination work in this corpus is a traceability-and-follow-through role situated between frontline observations, supervisor routines and specialist decisions. Its strongest recurring requirements are supervisor coordination, reliable records, field observation and incident intake. Hazard reporting, worker participation, corrective-action follow-through, safety meetings and escalation add important but source-sensitive elements. Good coordination preserves facts, keeps ownership visible, supports two-way communication and recognizes when a decision belongs to someone with different authority or expertise.

Selected public references

Archive and citation

The verified version of this research report is archived on Zenodo at DOI 10.5281/zenodo.23072469, with a direct public PDF. Technical report number: MTF-CF-RR-2026-10-01-57. Publication date: 1 October 2026.

Continue learning

Apply the evidence from this report through MTF Institute's Professional Certificate in Occupational Health and Safety Coordination. The programme turns the identified capabilities into structured theory, guided AI practice and reusable workplace artifacts.