ENGINEERING & PRODUCT DELIVERY · 9 NOVEMBER 2026
Building a personal AI system: architecture and delivery roadmap
Try the interactive public prototypes · Existing AI research laboratory · Research
The intended product is a modular assistant environment rather than one unrestricted system. Separate capabilities should share a private retrieval layer, a permission broker and an action history. Each component must be independently testable and replaceable.
1. Component architecture
| Service | Inputs and outputs | Implementation requirements |
|---|---|---|
| Identity and consent | User sign-in, scoped permissions, connected service tokens | Authentication, least privilege, credential rotation, encrypted secrets, revocation |
| Private file vault | Uploaded PDFs, notes, papers, metadata, retention preferences | Encrypted file storage, access enforcement, file type controls, deletion policy |
| Search and retrieval | Passage indexing, document IDs, source pages, citations | Hybrid lexical/vector retrieval, citation coverage tests, source-aware versioning |
| Model gateway | Questions, approved context, structured proposed outputs | Server-side API keys, provider abstraction, rate limits, budget caps, model evaluations |
| Agent workflow engine | Explicit task plans and allowed operations | Allowlisted tools, tool-level permissions, injection checks, isolated execution |
| Action ledger | Proposal, approval, execution status, external receipt, rollback attempt | Auditable events, idempotency, error recovery, clear irreversible-action notices |
| Connectors | Mail, calendar, content platforms, business systems | Official APIs, per-action scopes, usage limits, platform terms, consent |
| Monitoring | Errors, effectiveness, security incidents, evaluations | Minimised logs, anomaly alerting, security review and routine maintenance |
2. Phased delivery
| Stage | Demonstrable outcome | Exit criteria |
|---|---|---|
| 0 — Public interaction | Browser-only demos: evidence matching, triage, tutoring, content queue and replies | Accessible interface; no secret credentials; clear prototype labelling |
| 1 — Grounded research assistant | Private document upload and question answering with quoted passages and citations | Citation tests, permission isolation, deletion tests, measured error cases |
| 2 — Business process pilot | Human-approved intake and draft workflows with simulated and then limited real integrations | No unexpected external actions; rollback and failure state verified |
| 3 — Learning and communications | Personalised explanations and editable one- or two-option message drafts | Content quality tests, author approval, contextual correctness |
| 4 — Licensed content scheduling | Media library, queue, preview, approval and selected platform APIs | Publishing permissions, duplicate checks, platform compliance, stop/pause |
| 5 — Proactive personal assistant | Reminders and suggested actions from user-approved records | Opt-in triggers, clear reasons and sources, approvals for purchases, bookings or submissions |
3. Action permissions and reversibility
Classify operations as read-only, reversible local changes, consequential external changes or legally/financially consequential actions. Let an agent automatically gather permitted information and prepare a plan, but require explicit approval before sending correspondence, publishing to third-party accounts, paying, deleting records, booking paid appointments or filing official forms. Keep a visible undo button only for actions actually reversible; an email that was delivered cannot be reliably recalled.
4. Evaluation and scientific value
Each specialist assistant should have reproducible evaluation data: source-grounding precision, coverage, task completion, unsupported assertion rate, action safety, cost per successful task, latency, accessibility and privacy tests. Compare against a non-AI baseline and record cases where a human is more reliable. Publish sanitised benchmarks and research methods, not private personal data.
5. Service and commercial models
Candidate services: in-person AI configuration demonstrations; fixed-scope research retrieval setups; private workflow analysis; organisational training; maintenance agreements where there is an ongoing, measurable service. Test willingness to pay before adding recurring infrastructure costs. A prototype is not a commercial guarantee, and formal commitments should reflect actual delivery capability.
6. What is not live yet
There is currently no connected language-model backend, account-based sign-in, encrypted private vault, autonomous mail responder, official content scheduling integration or booking/purchase service attached to this demonstration. GitHub Pages publishes static content and cannot run server-side application code by itself. A separate protected backend and operating budget are required.
7. Security and standards references
GitHub Pages hosting limitations · OWASP risks for LLM applications · NIST AI Risk Management Framework · ICO AI and data protection resources
Development architecture and phased plan · Not a claim of deployed private services or production readiness.