AI Privacy Engineering Review
Trace personal data through your agent workflow. Examine collection, memory, retrieval, tool access and retention against your agreed requirements.
Make personal-data persistence and access inspectable
- Collection purpose
- Connect each data category to an intended workflow use.
- Persistent copy
- Trace prompts, memory, indexes and logs.
- Access boundary
- Check which identity can retrieve or transfer information.
- Deletion evidence
- Follow removal through the retained copies in scope.
Decision and fit
Make personal-data persistence and access inspectable
Compare the engagement with the decision you face
- A question worth resolving
- Make personal-data persistence and access inspectable.
- An accountable participant
- Privacy and engineering owners who need to understand personal-data flows through an agent's prompts, tools and persistent state.
- Inspectable source material
- The relevant owners can make workflow evidence and agreed inputs available for examination.
- A scope mismatch
- The primary requirement is a legal opinion or compliance attestation.
The engagement
Trace personal data through your agent workflow. Examine collection, memory, retrieval, tool access and retention against your agreed requirements. A technical data-flow and control review that supplies evidence for your privacy decision-makers.
The decision at stake
The team cannot explain where personal data persists or which agents and tools can retrieve it.
Suitable situations
- Privacy and engineering owners who need to understand personal-data flows through an agent's prompts, tools and persistent state.
Less suitable situations
- The primary requirement is a legal opinion or compliance attestation.
Engagement boundary
Agree the work and its boundaries together
The engagement boundary
- Collection purpose
- Trace personal data through prompts, memory, retrieval and tools.
- Persistent copy
- The inputs, authorised access and client participants are confirmed before scheduling.
- Privacy and engineering owners
- Retains organisational decisions and risk acceptance.
- Additional implementation
- Legal advice, regulator submissions and a formal privacy impact assessment opinion are outside the engineering review. Implementing controls and verifying later data-flow changes need separate agreement.
What the work covers
- Trace personal data through prompts, memory, retrieval and tools.
- Examine access, retention, deletion and external transfer controls.
- Document engineering gaps and questions requiring qualified privacy review.
Scope boundaries
- Formal certification, legal conclusions and guarantees of safe or compliant behaviour.
- Implementation or ongoing operation beyond the agreed scope.
- Legal advice, regulator submissions and a formal privacy impact assessment opinion are outside the engineering review. Implementing controls and verifying later data-flow changes need separate agreement.
Decision artefacts
Leave a record the next owner can use
Data-flow finding
Illustrative working format
- Subject and version
- Identify the workflow, configuration or proposal examined.
- Collection purpose
- Record the relevant requirement and supporting material.
- Access boundary
- State the observation, supporting evidence and remaining uncertainty.
- Privacy and engineering owners
- Record the next decision, follow-up and unresolved question.
What the work can produce
- Data flow and retention map
- Privacy control findings
- Engineering actions for specialist review
Working sequence
How the work moves from question to handover
What each stage should leave your team with
- Map data movement
Data categories, stores and provider transfers mapped.
- Inspect control behaviour
Access and lifecycle controls matched to supplied requirements.
- Exercise agreed lifecycle cases
Observed expiry, correction and deletion behaviour.
- Brief privacy and engineering owners
Engineering actions separated from legal questions.
How the work proceeds
- Map data movement. Trace personal data from collection through context, retrieval, tool calls, logs and memory, including external providers.
- Inspect control behaviour. Review minimisation, access, retention, correction and deletion against the purposes and requirements supplied by the client.
- Exercise agreed lifecycle cases. Use approved data to observe access changes, expiry and deletion propagation across selected stores.
- Brief privacy and engineering owners. Separate evidenced technical gaps from legal questions, and assign control actions with verification steps.
Worked situation
Follow one situation through to a bounded next decision
Illustrative working situation
Illustrative scenario
- Starting condition
A deleted source remains retrievable from a derived index.
- Question to examine
Inspect the propagation of deletion across the data flow.
- Possible next action
Record the engineering gap for qualified privacy review.
Working together
Your team and our team need a clear working agreement
Who contributes and who decides
- Client sponsor
- Confirms the decision, priorities and people needed to resolve it.
- Client technical or process owner
- Provides the agreed evidence, access and review of findings or deliverables.
- Softlander delivery lead
- Coordinates the agreed activities, review points, handover and open actions.
- Authorised decision-maker
- Accepts deliverables against the agreed criteria and separately decides operational or business action.
What the work depends on
- Engineering and privacy contacts provide data categories, processing purposes, supplier interfaces, retention settings and approved test records. Your qualified privacy adviser determines legal requirements and reviews questions beyond technical control behaviour.
Engagement availability
The start date, named delivery roles, review points and fee basis are agreed after confirming the inputs and scope. Duration depends on evidence access, system complexity and the time your decision-makers need to review the work.
Engagement and handover
Agree the delivery model and what completion means
A clear agreement from scope to ownership
- Scoped work
- Trace personal data through prompts, memory, retrieval and tools.
- Indicative handover
- Data flow and retention map; Privacy control findings
- Review and acceptance
- Review the agreed deliverables and record any open items, owners and conditions.
- Ongoing work
- Assign maintenance and operational ownership; agree additional scope and coverage explicitly.
A technical data-flow and control review that supplies evidence for your privacy decision-makers.
The handover links each observed data flow and retention or deletion gap to an affected component, supporting evidence and engineering action. Unresolved legal interpretations are referred to the client's qualified adviser rather than represented as conclusions.
Legal advice, regulator submissions and a formal privacy impact assessment opinion are outside the engineering review. Implementing controls and verifying later data-flow changes need separate agreement.