IndustryEvidence, not intent
Compliance evidence is the first automation
Statutory certificates and their expiry dates are a bounded, high-consequence document problem, which makes them the clearest early target in an estate.
Program 12 · Domain specialisation
SCIKIQ Certified Data and AI EngineerFacility Management
Facility management looks like a services business and behaves like a compliance one. A building has statutory obligations with legal dates attached — fire, lifts, water hygiene, electrical — and the question a regulator or an insurer asks is not whether you meant to comply but whether you can produce the certificate. That is a document and evidence problem, which is exactly what this specialisation is built around.
- 80 hrstaught
- 40hands-on AI labs
- 8assessment tests
- +20 hrs2 E2E capstones
Statutory floorObligations with legal dates that outrank every efficiency goal
EnergyThe largest controllable operating cost in most portfolios
Reactive mixMaintenance that is unplanned, and the cost that follows from it
OccupancySpace paid for and not used, measurable at last
01
Part 1 · The business
The value chain, the people, the trends
The estate is a portfolio of buildings, a book of service contracts and a legal obligation register. Part 1 maps all three and names the people holding them together.
The facilities value chain
Six stages from portfolio to occupant. Select one.
01
Portfolio & space
The estate, leases, space allocation, occupancy and the cost of holding it.
Decisions made here
- What do we hold?
- Is this space earning its cost?
- Where do we consolidate?
Where it breaks today
- Space data that disagrees with the lease and with reality
- Occupancy inferred from badge data of unknown quality
- Portfolio decisions made on floor area rather than on use
What is measured
- Cost per square metre
- Utilisation
- Vacancy
Where AI landsSpace and lease reconciliation, utilisation analysis and consolidation modelling.
Cost per square metreUtilisationVacancy
02
Plan & procure services
Service specifications, contractor selection, SLA structures and mobilisation.
Decisions made here
- What do we specify?
- Who delivers it?
- Is this contract performing?
Where it breaks today
- Service specifications inherited and never re-examined
- Contract obligations spread across documents nobody reads together
- Performance argued at the review rather than evidenced continuously
What is measured
- Contract cost
- SLA performance
- Variation spend
Where AI landsContract obligation extraction, SLA performance evidencing and variation review.
Contract costSLA performanceVariation spend
03
Operate & maintain
Hard and soft services, planned and reactive maintenance, helpdesk and dispatch.
Decisions made here
- What needs doing?
- Who do we send?
- Is this planned or has it already failed?
Where it breaks today
- Reactive work crowding out the planned programme
- Helpdesk tickets described by occupants, not by fault
- Asset history spread across systems and contractors
What is measured
- Planned vs reactive mix
- First-time fix
- Response time
Where AI landsTicket triage from occupant language, failure prediction and dispatch support.
Planned vs reactive mixFirst-time fixResponse time
04
Energy & sustainability
Consumption, plant efficiency, controls strategy, carbon reporting and retrofit.
Decisions made here
- Why is consumption up?
- Which plant is inefficient?
- Where does retrofit pay?
Where it breaks today
- Building management systems running strategies nobody documented
- Consumption anomalies found in the monthly bill
- Carbon reporting assembled by hand each period
What is measured
- Energy intensity
- Carbon emissions
- Plant efficiency
Where AI landsConsumption anomaly detection, control-strategy review and retrofit prioritisation.
Energy intensityCarbon emissionsPlant efficiency
05
Compliance & safety
Statutory inspections, certificates, permits, risk assessments and remedial actions.
Decisions made here
- Are we compliant?
- What expires next?
- Has the remedial action been closed?
Where it breaks today
- Certificates held by contractors rather than by the client
- Compliance status assembled for an audit rather than known continuously
- Remedial actions raised and never verifiably closed
What is measured
- Compliance coverage
- Overdue inspections
- Open remedials
Where AI landsCertificate extraction and expiry tracking, gap detection and remedial closure evidence.
Compliance coverageOverdue inspectionsOpen remedials
06
Serve the occupant
Workplace services, requests, environmental comfort and occupant experience.
Decisions made here
- What are occupants telling us?
- Is this space working?
- What do we fix first?
Where it breaks today
- Complaints handled individually rather than as a signal
- Comfort issues reported long after the condition started
- Experience measured by survey once a year
What is measured
- Occupant satisfaction
- Request resolution time
- Comfort complaints
Where AI landsComplaint clustering to building conditions and pre-emptive comfort intervention.
Occupant satisfactionRequest resolution timeComfort complaints
Who you are building for
One of these people is personally accountable if a statutory certificate has expired. Select one to light the stages they own.
What the sector is seeing
A services industry whose data finally exists, and whose obligations were always legal.
IndustryCompare to strategy
Building systems became readable
Modern controls and metering expose enough data that plant behaviour can be compared against the documented strategy rather than assumed to follow it.
IndustryPlanned beats reactive
Reactive mix is the cost lever
Unplanned work costs materially more than the same work planned, so shifting the mix is where operating savings in maintenance actually come from.
IndustryMeasured utilisation
Occupancy data made space a measurable cost
Utilisation can now be measured rather than surveyed, which turns portfolio decisions from floor-area arithmetic into evidence.
IndustryReported, not estimated
Carbon reporting became an obligation
Emissions reporting has moved from voluntary to expected, and assembling it by hand each period does not scale with the number of buildings.
IndustryAggregate by design
Occupant sensing raised a privacy question
The same data that measures utilisation can identify individuals, so aggregation and purpose limitation are design decisions rather than policy footnotes.
02
Part 2 · Data and meaning
Domains, ontology, taxonomy and governed terms
Two things make this architecture distinctive: building systems are operational technology with a real safety boundary, and the compliance obligations are held in certificates issued by third parties. Neither is a database you can simply query.
The data landscape
Six families. The certificate corpus is the one that carries legal consequence.
Asset & space register
Buildings, floors, spaces, plant and equipment records.
CAFM / IWMS, space systemsBuilding management systems
HVAC, lighting, controls, setpoints and plant telemetry.
BMS / BAS, meteringWork orders & helpdesk
Planned and reactive jobs, tickets, dispatch and completion.
CAFM, helpdesk platformsCompliance certificates
Statutory inspections, certificates, risk assessments and remedials.
Contractor portals, document storesLeases & service contracts
Lease terms, service specifications, SLAs and variations.
Lease systems, contract storesOccupancy & experience
Access events, sensor occupancy, bookings and occupant requests.
Access control, sensors, booking systemsThe semantic layer
A facilities agent that cannot tell an obligation from a certificate will report a building as compliant because a document exists, or as non-compliant because one is missing from a portal. These are the domains, the graph and the definitions that make the difference.
Data domains
Six subject areas, each with the entities it holds, its critical data elements and the function accountable for it.
Estate & space
The buildings and the space inside them.
PropertyBuildingFloorSpace
- Critical data elements
- Property identifier; Net internal area; Space use type
- Owned by
- Estates
Asset
The plant and equipment that has to keep working.
AssetAsset classSystemComponent
- Critical data elements
- Asset identifier; Asset class code; Installation date
- Owned by
- Engineering
Obligation & evidence
What the law requires, and what proves it was done.
ObligationInspectionCertificateRemedial action
- Critical data elements
- Obligation frequency; Inspection date; Expiry date
- Owned by
- Compliance
Work
What was asked for, who did it and whether it closed.
Work orderTaskContractorCompletion
- Critical data elements
- Priority; Response time; Completion status
- Owned by
- FM operations
Energy
What the estate consumed and what it emitted.
MeterConsumptionEmissionRetrofit
- Critical data elements
- Meter reading; Energy intensity; Emission factor
- Owned by
- Sustainability
Occupancy
Who used the space, aggregated so that nobody is identifiable.
Access eventOccupancy countBookingRequest
- Critical data elements
- Aggregated occupancy; Booking duration; Request type
- Owned by
- Workplace
Ontology & knowledge graph
The entities and the typed relationships between them — what the agent traverses instead of guessing joins. Select any entity.
Core entityEventReference / classification
Core entity
Building
A structure in the estate.
- Building contains Space
- Asset installed in Building
- Meter measures Building
Core entity
Space
An area within a building.
- Building contains Space
Core entity
Asset
Plant or equipment installed in the building.
- Asset installed in Building
- Asset classified as Asset class
- Work order raised against Asset
Reference / classification
Asset class
The classification that determines duties.
- Asset class carries Obligation
- Asset classified as Asset class
Reference / classification
Obligation
A statutory duty attaching to an asset class.
- Asset class carries Obligation
- Inspection satisfies Obligation
- Service contract specifies Obligation
Event
Inspection
The performance of a statutory check.
- Inspection satisfies Obligation
- Inspection evidenced by Certificate
- Inspection raises Remedial action
Event
Certificate
The evidence an inspection was passed.
- Inspection evidenced by Certificate
Event
Remedial action
Work an inspection finding requires.
- Remedial action closed by Work order
- Inspection raises Remedial action
Event
Work order
A job raised against an asset.
- Work order raised against Asset
- Remedial action closed by Work order
- Contractor performs Work order
Core entity
Contractor
The party performing the work.
- Contractor performs Work order
- Service contract governs Contractor
Core entity
Meter
A consumption measurement point.
- Meter measures Building
Reference / classification
Service contract
The agreement governing a contractor.
- Service contract governs Contractor
- Service contract specifies Obligation
Taxonomies
The classification hierarchies that make records comparable across systems.
Asset hierarchy
- Property
- Building
- System
- Asset
Campus North → Block B → Chilled water → Chiller 2
Statutory duty
- Regime
- Duty
- Frequency
Fire safety → Emergency lighting test → Monthly
Work classification
- Type
- Priority
- Trigger
Reactive → P1 → Loss of heating in occupied space
Governed business terms
The definitions an agent must use rather than invent. Most wrong answers in this industry are a term used loosely.
- Obligation vs certificate
- The obligation is the legal duty; the certificate is evidence it was met. Counting certificates you hold measures your filing, not your compliance.
- Missing vs absent
- A missing document may mean the inspection was never done, or that the contractor still holds the paperwork. Reporting them identically destroys trust in the report.
- Statutory compliance
- Duties imposed by law with dates attached, evidenced by certificate rather than by intention.
- Remedial action
- Work required by an inspection finding — the item most often raised and least often verifiably closed.
- Planned vs reactive
- Maintenance done to a schedule versus after failure. The mix, not the volume, is the cost lever.
- Hard vs soft services
- Engineering services versus cleaning, security and catering; they have different data and different contracts.
- Utilisation
- How much of the space paid for is actually used, measured rather than surveyed.
- Aggregation threshold
- The minimum grouping below which occupancy data is not reported, because below it the data identifies individuals.
03
Part 2 · The architecture
How it is actually built
The layers a deployment needs, what each one holds, and what the engineer owns there.
The reference architecture
Seven layers, with the statutory obligation register treated as a governed entity.
01/07
Estate, building & certificate sources
The asset and space register, building systems, work orders and the certificate corpus.
Takes in
Nothing. This is the upstream edge.
Hands on
Asset-resolved telemetry, extracted certificates with expiry dates, and aggregated occupancy.
SystemWhat it doesKey business entities
CAFM / IWMS
The facilities system of record: assets, locations, planned maintenance, work orders and space. Everything else has to resolve back to its asset identifiers.
AssetLocationWork orderSpace
Building management system
HVAC, lighting and plant controls with their setpoints and schedules. Operational technology, read across a controlled path and never written to.
PointSetpointScheduleAlarm
Metering and sub-metering
Electricity, gas, water and heat consumption at building and sometimes circuit level. The basis of every energy and carbon number.
Meter readingConsumptionCarbon
Compliance certificate stores
Statutory inspection certificates, risk assessments and remedial actions — frequently held in contractor portals rather than by the client who is legally accountable.
CertificateInspectionExpiryRemedial
Service contracts and leases
Service specifications, SLAs, variations and the lease obligations that sit under the estate.
ContractSLAObligationLease
Occupancy and access
Badge events, sensor occupancy and room bookings. Personal data at fine resolution, which is why aggregation is applied before anything else reads it.
Access eventOccupancyBooking
What the engineer owns here
Getting certificates out of contractor portals and into the client’s own evidence base.
Where this layer fails
Reporting a compliance gap that is really a missing document. The certificate usually exists and sits in a contractor portal, and a system that cannot tell the two apart destroys its own credibility on the first report.
02/07
Ingestion & asset context
Telemetry and work-order ingestion aligned to the asset hierarchy, plus certificate extraction.
Takes in
Read-mostly access paths, change feeds and document streams from the systems of record.
Hands on
Replayable, resolved, quality-checked records with their lineage back to the source row.
SystemWhat it doesKey business entities
Change data capture
Reads the source’s own change log rather than polling it, so the platform sees every state a record passed through instead of only where it ended up.
Change eventOperation typeCommit timestamp
Landing and raw zone
Every extract kept as it arrived, immutable and timestamped. The thing you replay from when a downstream definition turns out to have been wrong.
Raw extractBatchWatermark
Document processing
Layout-aware extraction over the unstructured half of the estate: sectioning, tables, signatures, and the page reference every later citation depends on.
DocumentSectionPage reference
Entity resolution
Decides that two records are the same real-world thing, with a survivorship rule and a confidence, so downstream joins are a decision rather than an assumption.
Match candidateSurvivorship ruleCluster
Data quality and contracts
Schema, freshness and volume expectations asserted at the boundary, so a bad load fails loudly here rather than quietly two layers later.
ContractExpectationViolation
Certificate extractionthis domain adds
Pulls asset, inspection type, date and expiry out of certificates issued in whatever format each contractor happens to use.
CertificateInspectionExpiry
What the engineer owns here
Resolving a BMS point and a work order to the same physical asset.
Where this layer fails
Treating an unreadable scan as an absent certificate. The obligation may well have been met; what failed was your extraction, and reporting it as a compliance gap is worse than silence.
03/07
Obligation model & semantics
The graph joining building, asset, obligation, certificate, contract and responsible party.
Takes in
Replayable, resolved records with lineage from ingestion.
Hands on
Typed, classified, defined data an agent can be grounded on without inferring meaning.
SystemWhat it doesKey business entities
Business glossary
One agreed definition per term, owned by a named person, so the number an agent quotes means what the business means by it.
TermDefinitionOwnerSteward
Ontology and knowledge graph
The typed entities and relationships the domain actually has, so an agent can traverse "which customers are exposed to this" rather than guess from adjacent text.
EntityRelationshipNodeEdge
Semantic and metrics layer
Metrics defined once, in one place, with their filters and grain. Removes the class of error where the agent computed something plausible and wrong.
MetricDimensionGrainFilter
Classification and policy
Sensitivity labels and the purpose each classification permits, applied at the field level and inherited by everything downstream.
ClassificationPurposePolicy
Lineage and catalogue
Where a value came from and what it touched on the way. The layer that makes an answer defensible rather than merely correct.
Lineage edgeDatasetColumn
Statutory obligation registerthis domain adds
States which assets carry which legal duty, at what frequency and under whose ownership. Compliance coverage has no meaning without it.
ObligationAssetFrequencyOwner
What the engineer owns here
One register stating which assets carry which statutory obligation, and who owns each.
Where this layer fails
Measuring compliance against the certificates you happen to hold rather than against the obligations that exist. That produces a reassuring number and no information.
04/07
Retrieval & grounding
Retrieval over regulations, standards, service contracts, O&M manuals and risk assessments.
Takes in
Typed, classified, defined data from the governance layer.
Hands on
Ranked, permission-trimmed, citable evidence scoped to the caller and the moment.
SystemWhat it doesKey business entities
Hybrid index
Vector and lexical retrieval together, because exact identifiers, codes and clause numbers are the thing semantic search is worst at.
ChunkEmbeddingTermScore
Chunking and metadata
Splits on the document’s own structure and carries effective dates, version and source into every chunk, so a retrieved passage knows when it was true.
ChunkEffective dateVersionSource reference
Security trimming
Applies the caller’s permissions inside the query rather than filtering results afterwards, so the model never sees what the user may not.
PrincipalEntitlementFilter
Graph retrieval
Traverses the knowledge graph for questions that are joins rather than similarity — exposure, genealogy, ownership, causation.
PathHopTraversal
Reranking and citation
Reorders candidates on relevance and returns the passage identifier behind every sentence, so the answer can be checked rather than trusted.
CandidateRankCitation
What the engineer owns here
Answering an obligation question with the regulation and the contract clause cited.
Where this layer fails
Filtering results after retrieval instead of constraining the query. The model has already seen the rows you removed, and it will use them.
05/07
Agent & orchestration
Agents for compliance gap detection, ticket triage, anomaly review and contract evidencing.
Takes in
Ranked, permission-trimmed, citable evidence from retrieval.
Hands on
Actions taken or proposed, each with the evidence, the identity and the trace behind it.
SystemWhat it doesKey business entities
Agent runtime
Plans, calls tools and holds the loop. Where the step budget, the timeout and the stopping condition are enforced rather than hoped for.
TaskPlanStepTrace
Tool and MCP servers
Typed, permissioned tools with declared schemas. An agent’s real capability surface is this list, which is why the list is a security artefact.
ToolSchemaScopeInvocation
State and memory
Durable task state with checkpoints, so a run that dies mid-way resumes instead of restarting and re-doing side effects.
CheckpointSessionMemory record
Human-in-the-loop gates
Approval steps on the actions that need one, carrying enough context for the approver to actually decide rather than rubber-stamp.
ApprovalThresholdDecision
Workflow integration
Writes back into the systems people already work in, under a service identity with its own audit trail.
Work itemWrite-backActor
What the engineer owns here
Keeping the agent advisory on anything that touches building control.
Where this layer fails
An agent that acts under a service account rather than on behalf of the user. It will eventually do something the requesting user had no right to do, and the log will not show it.
06/07
Safety, privacy & guardrails
The building-controls boundary, occupant privacy, aggregation thresholds and audit.
Takes in
Proposed actions and drafted responses from the agent layer.
Hands on
Permitted, grounded, logged output — or a refusal with a stated reason.
SystemWhat it doesKey business entities
Policy enforcement
The rules that decide whether an action is permitted at all, evaluated before the action and independently of the model that proposed it.
PolicyDecisionObligation
Input and output screening
Injection detection on the way in, and on the way out the checks for leakage, unsupported claims and content the domain forbids.
PromptResponseViolation
Groundedness checking
Verifies each assertion resolves to retrieved evidence, and fails the response rather than shipping the sentence that does not.
AssertionSupportVerdict
Model risk management
Model inventory, intended use, validation evidence and the sign-off that lets a model be used for a purpose. Not optional in a regulated estate.
ModelIntended useValidationApproval
Audit log
Immutable record of what was asked, retrieved, decided and done — the artefact a reviewer reads when they do not take your word for it.
Audit entryActorEvidence reference
Occupancy aggregationthis domain adds
Enforces the minimum aggregation below which utilisation output is not returned, because at fine resolution it identifies individuals.
ThresholdAggregateRe-identification test
What the engineer owns here
Proving occupancy analysis cannot identify an individual, and control cannot be actuated.
Where this layer fails
Utilisation reporting at desk-and-hour resolution. It is personal data about named staff, whatever the dashboard is called.
07/07
Evaluation & observability
Compliance coverage, prediction precision, anomaly accuracy and cost per building.
Takes in
Permitted, grounded, logged output from the guardrail layer.
Hands on
Measured quality, cost and latency — and the evidence to change any of the three.
SystemWhat it doesKey business entities
Evaluation harness
Held-out sets and graded runs on every change, so a prompt edit is a measured change rather than a hopeful one.
Eval setCaseScoreBaseline
Regression gate
Blocks a release when quality drops, in CI, on the same evidence for everyone. The difference between a system and a demo.
ThresholdRunGate decision
Tracing
End-to-end spans across retrieval, model and tool calls, so a bad answer can be opened and read rather than argued about.
TraceSpanLatencyToken count
Cost and capacity
Cost per task and per tenant, against throughput and quota. The number that decides whether the pilot can become the rollout.
Cost per taskQuotaThroughput
Drift and feedback
Watches quality against production traffic rather than the test set, and routes real corrections back into the eval suite.
Drift signalFeedbackIncident
What the engineer owns here
Proving compliance coverage improved, which is the only number an auditor cares about.
Where this layer fails
Evaluating once, before launch. Quality moves with the data, the model and the traffic, and a system with no live measurement has no idea which of the three moved.
Typical tech stack
Models
- Frontier models for certificate and contract reasoning
- Small models for ticket triage
- Anomaly models on consumption
Grounding
- Regulation and standards index
- Service contract clause index
- Asset knowledge graph
Agents
- Compliance gap agents
- MCP over CAFM and BMS reads
- Ticket triage
Data
- Certificate extraction with expiry dates
- BMS point-to-asset resolution
- Occupancy aggregation
Assurance
- Compliance coverage measurement
- Anomaly precision evals
- Privacy aggregation tests
Runtime
- Read-only BMS path
- Multi-tenant portfolio isolation
- Cost per building budgeting
Constraints that shape the design
Statutory obligations set a floor
Fire, lifts, water hygiene and electrical inspections have legal dates. No efficiency gain justifies a gap, and the evidence is the certificate rather than the intention.
Building controls are operational technology
The agent reads the BMS and never writes to it. Comfort and safety strategies are engineered artefacts with their own change control.
Occupancy data identifies people
At fine enough resolution, utilisation data is personal data. Aggregation thresholds have to be architectural, not a reporting convention.
The evidence sits with contractors
Certificates are often held by whoever performed the inspection. Any compliance system that assumes the client already has the documents will report a gap that is really a retrieval failure.
04
Part 3 · Getting it done
Where the business and the technology meet
The compliance use cases come first here, not because they are the most interesting but because everything else is optional and they are not.
Use cases, mapped to the chain and the architecture
Filter by stage or by earned autonomy. Selecting a use case jumps the value chain and the architecture to the stage and layer it depends on.
Value chain stage
Autonomy earned
12 use cases
No use cases match that combination.
The delivery playbook
How a facilities deployment actually runs.
01
Week 1 — build the obligation register
Establish which assets carry which statutory obligation and who owns each. Without this the compliance work has nothing to measure coverage against.
02
Week 2 — get the certificates in
Retrieve certificates from contractor portals into the client’s own evidence base. Reporting a gap that is really a missing document destroys trust immediately.
03
Weeks 3-6 — compliance first, everything else second
Certificate extraction and gap detection carry legal consequence and no controls risk. They also build the asset resolution the energy work will depend on.
04
Weeks 7-9 — energy and maintenance on a resolved asset model
Only once BMS points and work orders resolve to the same assets can anomaly detection and prediction mean anything.
05
Weeks 10-12 — privacy, controls boundary and handover
Prove occupancy analysis cannot identify an individual and that no control path exists, then hand over runbook, obligation register and coverage evidence.
The curriculum
8 modules, 80 taught hours, 40 hands-on labs and 8 assessments — every lab provisioned and graded by the SCIKIQ Agentic AI Playground. Open a module to see its labs.
01The SCDAI delivery model in facilitiesLanding in an operation where statutory obligation outranks every efficiency goal.10 hrs · 5 labs
Delivery mandateStatutory obligation basicsValue sizing on reactive and energyStakeholder mapContractor dynamicsThin-slice scoping
01
Decision selectionScore four facilities decisions on value, risk and data readiness.
02
Value sizingSize a reactive call-out, an energy anomaly and a compliance gap.
03
Register auditAssess whether the asset register supports an obligation register.
04
Stakeholder mapMap FM, compliance, hard services, energy and the contractor.
05
Thin sliceScope a single-building compliance slice with a measurable outcome.
Assessment 01 — Operating model & obligations
02Built-environment data foundationsRegisters, building systems, work orders, certificates and occupancy.10 hrs · 5 labs
Asset and space registersBMS and metering dataWork-order dataCertificate corporaOccupancy dataPoint-to-asset resolution
01
Register ingestionIngest an asset and space register and measure its quality.
02
BMS ingestionIngest building telemetry through a read-only path.
03
Point resolutionResolve BMS points and work orders to the same physical asset.
04
Certificate ingestionIngest scanned certificates with per-page quality scoring.
05
Occupancy aggregationAggregate occupancy so no individual can be identified.
Assessment 02 — Facilities data landscape
03Context engineering for buildings and obligationsAsset references, units, dates and the precision a legal obligation demands.10 hrs · 5 labs
Asset reference resolutionDate and expiry handlingUnits and conversionsStructured outputsUncertainty on poor scansContext budget
01
Asset resolutionResolve ambiguous asset references across register and BMS.
02
Expiry handlingExtract inspection and expiry dates with the calendar rules right.
03
Unit safetyPrevent energy unit-conversion errors in extracted readings.
04
Structured certificatesEmit certificate data as a typed, validated structure.
05
Context budgetCut context cost on a portfolio-wide extraction run.
Assessment 03 — Context engineering
04Grounding on regulation, contracts and manualsThe obligation is written down somewhere. Usually in three places that disagree.10 hrs · 5 labs
Regulation retrievalStandards and approved codesContract clause retrievalAsset graph traversalCitationsRetrieval evals
01
Regulation retrievalAnswer an obligation question citing the regulation and clause.
02
Contract clause retrievalFind what the service contract obliges for that asset class.
03
Asset traversalTraverse building to asset to obligation to certificate.
04
Citation contractCite regulation, contract clause and certificate on every answer.
05
Retrieval evalsBuild a compliance-officer-labelled question set and measure recall.
Assessment 04 — Retrieval & grounding
05Compliance and operations agent designGap detection where a false clean result is the dangerous failure.10 hrs · 5 labs
Workflow vs agentGap detection patternsMissing vs absentTicket triageAdvisory limits on controlsEscalation
01
Gap detection agentFind assets with no valid in-date certificate.
02
Missing vs absentDistinguish a missing document from a missing inspection.
03
Triage agentMap occupant complaint language to asset and fault class.
04
Advisory enforcementProve at the tool layer that no BMS write path is reachable.
05
EscalationDefine when a compliance finding must escalate and prove it does.
Assessment 05 — Agent design for compliance & operations
06Integrating with CAFM, BMS and contractorsRead the building, write to the work order, and never to the controller.10 hrs · 5 labs
MCP over CAFMBMS read pathsContractor portal integrationWork-order write-backIdentityError handling
01
MCP over CAFMExpose asset, work-order and certificate lookups as typed read tools.
02
BMS read pathDesign and prove a read-only path to building telemetry.
03
Portal retrievalRetrieve certificates from a contractor portal reliably.
04
Work-order write-backRaise remedial work orders under a traceable identity.
05
IdentityBind agent access to building, client and role entitlements.
Assessment 06 — Systems integration
07Safety boundary, privacy and securityTwo hard boundaries: the controller you must not touch and the occupant you must not identify.10 hrs · 5 labs
Building controls boundaryOccupant privacyAggregation thresholdsOWASP LLM Top 10Injection via scanned certificatesIncident response
01
Controls boundaryDocument and prove the agent sits outside building control.
02
Privacy thresholdsSet and test aggregation thresholds that prevent identification.
03
Re-identification testAttempt to identify an individual from utilisation output.
04
Injection defenceDefend against instructions embedded in a scanned certificate.
05
Red teamRun an OWASP LLM Top 10 pass over the compliance pipeline.
Assessment 07 — Safety, privacy & security
08Production, coverage and cost per buildingCompliance coverage is the number. Everything else is a supporting metric.10 hrs · 5 labs
Eval suitesCoverage measurementDriftFalse-clean detectionCost per buildingHandover
01
Eval gateGate deployment on certificate extraction accuracy regression.
02
Coverage studyMeasure compliance coverage before and after, honestly.
03
False-clean auditAudit specifically for assets wrongly reported as compliant.
04
Cost per buildingMeasure and reduce cost per building across the portfolio.
05
HandoverProduce runbook, obligation register and coverage evidence.
Assessment 08 — Production & coverage
Two capstones — 20 hours
End-to-end agent design, development, deployment and testing, 10 hours each. Reviewed by a Senior SCDAI Engineer against a published rubric.
Capstone A — Statutory compliance agent
Build an agent that extracts certificates across contractor formats, resolves each to an asset in the obligation register, and reports every asset without a valid in-date certificate — distinguishing a missing document from a missing inspection, and audited specifically for assets wrongly reported as compliant.
Capstone B — Consumption anomaly agent
Build an agent that detects departures from a building’s own weather- and occupancy-adjusted consumption pattern, cites the documented control strategy the plant appears to be departing from, and demonstrably has no write path to the BMS.
What you deliver, and how it is marked
- A running agent, deployed on the program's target stack, reachable by a reviewer.
- An architecture note: boundaries, tools, autonomy level and the trade-offs you took.
- A versioned eval set with a baseline and your final scores.
- A trace walkthrough of one success and one deliberate failure.
- Cost per task, measured — not estimated.
- A runbook and rollback path a client team could operate on Monday.
Problem fit15%Is this a decision worth automating, scoped to a defensible boundary?
Architecture20%Right pattern for the constraints; autonomy earned, not assumed.
Grounding & accuracy20%Measured groundedness against the eval set, with the failures named.
Safety & compliance20%Domain controls actually enforced, and demonstrably so.
Operability15%Traces, runbook, rollback, cost — could someone else run this?
Communication10%Can a client executive follow the reasoning in ten minutes?
One lab, start to finish
Every lab in this program follows the shape below. This is lab 05 in full — the brief you are given, the environment that is provisioned for you, the code you start from and the assertions that decide whether you passed.
Module 05 · Compliance and operations agent design
Compliance gaps that are real
Report every asset without valid statutory cover, and never confuse that with a missing file.
The brief
You are given an obligation register for 4,200 assets, certificates from six contractors in six different formats, and a labelled key. Some assets genuinely have no valid inspection. Others have one, evidenced by a certificate that is badly scanned or sitting in a portal you did not read. Report the first group. Reporting the second as non-compliant is the failure this lab is built to catch.
Provisioned for you
- 4,200 assets with statutory obligations
- Certificates from 6 contractors, 6 formats
- Deliberately degraded scans on 180 certificates
- Labelled key: true gaps vs retrieval failures
You start from this
def compliance_gaps(register, certificates):
"""Return [(asset, obligation, status, evidence)].
status is one of: 'compliant', 'gap', 'unverified'.
'unverified' is not a soft 'gap' -- it means the extraction failed and
a human must look, which is a different action entirely.
"""
# 1. an obligation is satisfied by an IN-DATE certificate for that asset
# 2. an unreadable certificate is unverified, never a gap
# 3. carry the evidence so a compliance officer can check any verdict
raise NotImplementedErrorWhat the grader asserts
- Identifies at least 95% of the labelled true gaps
- Reports zero true gaps as compliant — a false clean is the failure that matters here
- Degraded scans are returned as unverified, never as gaps
- Expiry is evaluated against the obligation frequency, not against the certificate date alone
- Every verdict carries the certificate reference and page that produced it
05
Assessment
Assessed, not attended
Every module ends with a timed, randomised assessment delivered through the SCIKIQ Agentic AI Playground. The certificate requires a pass on all of them plus two reviewed capstones.
Assessment blueprint
What each test covers, how long it runs, and how many items are currently in the versioned bank behind it.
| # | Assessment & coverage | Items | Time | In bank |
|---|---|---|---|---|
| 01 | Operating model & obligationsDelivery mandateStatutory obligationsValue sizingThin slicing | 25 | 35 min | 3 |
| 02 | Facilities data landscapeRegistersBMS dataCertificatesOccupancy and privacy | 25 | 35 min | 2 |
| 03 | Context engineeringAsset resolutionDates and expiryUnitsStructured outputs | 25 | 35 min | 2 |
| 04 | Retrieval & groundingRegulation retrievalContract clausesGraph traversalRetrieval metrics | 25 | 40 min | 2 |
| 05 | Agent design for compliance & operationsGap detectionMissing vs absentTriageAdvisory limits | 25 | 40 min | 2 |
| 06 | Systems integrationMCP designBMS read pathsWrite-back safetyIdentity | 25 | 35 min | 2 |
| 07 | Safety, privacy & securityControls boundaryOccupant privacyAggregationLLM security | 30 | 45 min | 2 |
| 08 | Production & coverageEval gatingCoverageFalse-clean riskEconomics | 30 | 45 min | 2 |
Single best answerOne defensible option among plausible distractors drawn from real field mistakes.
Scenario judgementA client situation with constraints; you pick the action a Senior SCDAI Engineer would take.
Architecture selectionGiven non-functional requirements, choose the pattern and justify the trade-off.
Failure diagnosisA trace, an eval report or a cost curve — identify the root cause.
- Each module ends with a timed assessment; the program certificate requires a pass on every one.
- Items are drawn at random from a versioned bank, so no two learners sit an identical paper.
- Pass mark is 70%. Two retakes are included, each with a fresh draw.
- Scenario items carry double weight — they are the ones that predict field performance.
- Capstones are reviewed against a published rubric by a Senior SCDAI Engineer, not auto-graded.
Try four questions from the bank
Real items, drawn from 17 in this program's bank — weighted toward the scenario and diagnosis types, because those are the ones that predict field performance. Instant feedback, nothing saved.
Four sample items — one attempt each, then the reasoning is shown.
Q1An FM provider asks for AI to reduce cost. Where should the first engagement land?
Compliance is the obligation that outranks efficiency, is a bounded document problem, and produces the asset resolution the energy work later depends on.
Q2An asset has an obligation and no certificate in your store. What is the correct status?
A missing document and a missing inspection require different actions. Reporting them identically destroys the credibility of the whole report on the first run.
Q3Where does the authoritative statement of a statutory duty come from?
The contract allocates the work; the law creates the duty. A compliance system grounded only in the contract will miss duties nobody contracted for.
Q4Which error matters most in compliance gap detection?
A false gap wastes time; a false clean means an unsafe asset sits unexamined with a report saying otherwise. Only one of those has legal consequence.
06
Career & outcomes
What this makes you, and where it goes next
The same ladder whichever specialisation you enter through — what changes is the domain you go deep in. Below: how the program is delivered, the skills it moves, the roles it leads to, and the specialisations closest to this one.
Available two ways
The same labs, assessments and capstones, delivered to an enterprise cohort or to individual professionals.
B2B
Direct to enterprise
Cohorts of 20 to 2,000+ on your own tenancy, with your data patterns and your cloud. Skill-gap baselining up front, per-team mastery reporting throughout, and capstones scoped against your real backlog so the output is deployable work.
- Private cohort on your tenancy
- Baseline skills assessment before kick-off
- Capstones scoped to your backlog
- Per-team mastery dashboards for L&D
- Optional Senior SCDAI Engineer architecture reviews
B2C
Direct to individual professionals
The same labs, assessments and capstones for individual engineers and analysts, run on shared infrastructure with a fixed cohort calendar. You leave with a graded portfolio, not a certificate of attendance.
- Fixed-date cohorts, self-paced within them
- Same graded labs and item bank
- Portfolio-ready capstones you own
- Certificate on a pass in every module
- Community and office-hours access
Skill matrix
Find your row and aim one column right. The Playground scores you against this after every module.
| Skill | Beginner | Intermediate | Advanced |
|---|---|---|---|
| Facilities domain fluency | Knows hard and soft services. | Maps obligations to assets and owners. | Sizes reactive, energy and compliance value credibly. |
| Obligation engineering | Reads a certificate. | Maintains an obligation register with coverage measured. | Audits for false-clean results, not just for gaps. |
| Building data | Reads BMS telemetry. | Resolves points and work orders to one asset model. | Designs a read-only path with privacy thresholds enforced. |
| Context engineering | Writes clear prompts. | Structures retrieval, tools and state deliberately. | Designs context strategy for reliability and cost at scale. |
| Retrieval & grounding | Builds basic vector search. | Tunes chunking, hybrid search and reranking. | Designs graph + vector grounding with measured recall. |
| Agent orchestration | Runs a single tool-calling agent. | Builds supervised multi-step and multi-agent flows. | Designs autonomy boundaries and failure containment. |
| Tool & system integration | Calls a documented API. | Writes an MCP server over a system of record. | Designs a least-privilege tool estate across systems. |
| Evaluation | Eyeballs outputs. | Builds labelled eval sets and regression gates. | Runs online evals with drift and judge calibration. |
| Observability & cost | Reads logs. | Traces runs, tracks tokens and latency. | Owns cost per task and capacity planning in production. |
| Security & guardrails | Adds output filters. | Mitigates the OWASP LLM Top 10 in a build. | Threat-models an agent estate and proves controls. |
| Client delivery | Takes notes in a workshop. | Runs discovery and scopes a thin slice. | Owns the account technically, from scope to handover. |
Where it takes you
The SCDAI ladder is the same whichever specialisation you enter through — what changes is the domain you go deep in.
01
Data and AI EngineerEntry
Python, LLM APIs, first RAG build, prompt and context basics.
02
Associate Data and AI EngineerPractitioner
Client-facing discovery, tool integration, agent build under supervision.
03
SCIKIQ Certified Data and AI EngineerSpecialist
Owns a deployment end to end — scoping, build, evals, handover.
04
Senior Data and AI Engineer / Domain LeadSenior
Multi-agent architecture, regulated deployments, reference patterns for the practice.
05
AI Architect / Practice LeadLeadership
Portfolio of accounts, delivery standards, hiring and capability strategy.
Terms worth knowing
- CAFM / IWMS
- The systems holding assets, work orders and space — the FM system of record.
- BMS / BAS
- Building management or automation system — the controls running plant.
- Statutory compliance
- Obligations imposed by law, with dates, evidenced by certificate.
- Planned vs reactive
- Maintenance done to a schedule versus after failure; the core cost mix.
- Remedial action
- Work required by an inspection finding, and the thing most often left open.
- Hard vs soft services
- Engineering services versus cleaning, security and catering.
- Utilisation
- How much of the space paid for is actually used, measured rather than surveyed.
- Obligation register
- The record of which assets carry which statutory duty, and who owns each.
- SCDAI
- SCIKIQ Certified Data and AI Engineer — the credential this program awards.
- SCDAI engineer
- An engineer embedded with the customer who turns their constraints into a working system.
- Thin slice
- The narrowest end-to-end path through a workflow that still produces business value.
- Agentic RAG
- Retrieval where the agent decides when, what and how often to retrieve.
- MCP
- Model Context Protocol — open standard connecting an agent to tools and data.
- A2A
- Agent2Agent — protocol for agent-to-agent delegation across vendors.
- Eval set
- A labelled, versioned set of cases an agent must pass before release.
- LLM-as-judge
- Using a model to score outputs, calibrated periodically against human ratings.
- Guardrail
- An enforced constraint on an input, output or action — not a prompt request.
- Human-in-the-loop
- A required human approval step before a consequential action executes.
- Cost per task
- Total token, tool and compute cost of one completed unit of work — the engineer's unit economic.
- Golden path
- The reference architecture a practice standardises on so deployments stay reviewable.
- Handover
- The point at which the client can operate, evaluate and change the system without the engineer.
Related programs
Skill a team, or join a cohort
B2B cohorts run on your tenancy with capstones scoped to your backlog. B2C cohorts run on a fixed calendar.
07
Enrolling
What you need, what you get, and what it costs
Stated plainly enough to rule yourself in or out without a sales call: the prerequisites, how the program runs, exactly what the credential is worth, and the questions everyone asks.
Before you start
Stated plainly so you can rule yourself in or out without a sales call. Nothing here is a formal qualification — it is what the first lab assumes you can already do.
Technical
You should already be able to do these
- Comfortable in Python: functions, typing, virtual environments, reading a stack trace
- Have called an LLM API and handled a structured response
- Git, the command line, and reading someone else’s code without panic
Domain
What we assume, and what we teach
- No building-services background needed; statutory duties are taught from scratch
- Readiness to treat a scanned certificate as the evidence that actually matters
Time and access
What the program asks of your week
- A machine that can run a browser and a terminal — labs run in the Playground, not locally
- Roughly 8 hours a week for the taught weeks, plus 20 hours for the two capstones
How it runs
Cohort dates and pricing are confirmed on enquiry rather than printed here, because both move with the intake.
- Format
- Self-paced within a fixed cohort window, delivered in the SCIKIQ Agentic AI Playground
- Taught hours
- 80 hours across 8 modules
- Capstones
- 20 hours — two end-to-end builds, reviewed not auto-graded
- Typical duration
- 10–12 weeks at around 8 hours a week
- Level
- Intermediate → Advanced
- Built for
- Engineers and analysts deploying AI across facilities and building operations
- Next cohorts
- — confirmed on enquiry
- Pricing
- — confirmed on enquiry
The credential
The credential is awarded per specialisation, so it names the domain or stack you were assessed in rather than claiming general competence. On this program the badge reads SCDAI — Facility Management.
How it is earned
- A pass at 70% or above on every module assessment — not an average across them
- Both capstones reviewed against the published rubric by a Senior SCDAI Engineer
- All labs submitted and machine-graded green
- Awarded as
- A verifiable digital badge with a public credential record
- Valid for
- Two years from award
- Recertification
- One current-year assessment plus a refreshed capstone
- Retakes
- Two included per assessment, each with a fresh item draw
- Named specialisation
- The badge states the domain or stack, e.g. SCDAI — Healthcare
- Verification
- Third parties confirm a holder from the credential record, no login required
A certificate that cannot be checked is decoration. Every award resolves to a record showing the specialisation, the award date and the assessments passed.
The certificate you leave with
This is the credential the program awards, shown exactly as it is issued — with the specialisation named, the assessment record attached and a verification link anyone can check without an account.
SCIKIQ Data Academy
Certificate of Professional Certification
This is to certify that
Your name
has been assessed and certified as
SCIKIQ Certified Data and AI Engineer
Facility Management
- Specialisation
- Facility Management
- Assessments passed
- 8 of 8, each at 70% or above
- Labs graded
- 40 of 40 machine-graded green
- Capstones reviewed
- 2, against the published rubric
- Taught hours
- 80 across 8 modules, plus 20 capstone hours
AwardedOn completion
Valid untilTwo years from award
Credential IDSCDAI-FACILITY-MAN-<issued>
Senior SCDAI Engineer, Certification Board
- 01The specialisation is on the certificate
Not “Data and AI Engineer” but the domain or stack you were actually assessed in. A general claim would be a weaker one.
- 02The assessment record travels with it
Modules passed, labs graded and both capstones reviewed — so the credential states what was measured rather than that you attended.
- 03Anyone can verify it, without a login
The credential ID resolves to a public record showing the specialisation, the award date and the assessments passed.
Share it on LinkedIn
Add it to your LinkedIn profile in one step. The link pre-fills the certification fields from the credential record, so the entry on your profile matches the record a reader can check.
Add to LinkedIn profile The button is live on your real certificate; here it opens LinkedIn pre-filled with this specialisation so you can see exactly what the profile entry will say.A certificate that cannot be checked is decoration. Every award resolves to a record showing the specialisation, the award date and the assessments passed.
Questions we get asked
The objections that come up in every conversation about this program, answered without the brochure voice.
Is this a course or a certification?
Both, and the second is the point. Eight timed assessments and two reviewed capstones stand between you and the credential, so a pass means someone measured the skill rather than recorded your attendance.
How is it self-paced without falling apart?
Every lab is provisioned, graded and unblocked by the Playground rather than by an instructor. That is what lets a cohort of 2,000 cost the same faculty time as a cohort of 20 — and why you are never waiting on someone to mark your work.
What if I fail an assessment?
Two retakes are included per assessment, each drawing a fresh item set from the bank, so a retake is a genuinely new paper rather than the same questions again.
Do I need my own cloud account?
For the domain specialisations, no — labs run in provisioned sandboxes. For the four tech-stack programs you will want access to that platform, since deploying into a real subscription is much of the point.
Can we run this privately for our team?
Yes. B2B cohorts run on your own tenancy with your data patterns, a skills baseline before kick-off, per-team mastery reporting, and capstones scoped against your actual backlog so the output is deployable work rather than an exercise.
Which specialisation should I take?
Take the domain you deploy into. If you move across industries, take a tech-stack program instead and pick up domain context on the engagement. The chooser on the programs page will narrow it.
How current is the material?
The trends, platform capabilities and regulatory positions are reviewed each quarter, and every external claim on these pages links to its source so you can check the date yourself.
What do I actually leave with?
A graded portfolio: forty machine-graded labs, two reviewed end-to-end agent builds with measured evaluation and cost per task, and a verifiable credential naming your specialisation.
Still deciding?
Tell us the systems you deploy into and we will say plainly whether this specialisation is the right one — or which of the 21 is.