src/data/facts.json · 51 entries
Fact file
Every claim on this site, in one place. Site copy is written from this file, and the role-match agent may only state things that trace back to an entry here. That includes the scope: what the work does not cover.
Person
on the home page →| p1 | Andrew Deighan. British and Irish citizen with full UK and EU right to work, relocating to the UK. Open to roles in the UK, Ireland, the EU and remote. Available from mid-October 2026. |
| p2 | Co-founder and Product Lead (CEO) of AtlasOra, June 2024 to October 2026. Two-sided vacation rental marketplace. Led a team of thirteen. |
| p3 | Founder of DUDsJobs, an AI job-hunting service, live with paying users since September 2026. |
| p4 | Head of Marketing at Qubic, an AI-focused layer-1 network, August 2024 to May 2025. |
| p5 | Nine years in secondary education across the UK, Thailand, New Zealand and Jordan, finishing as Assistant Headteacher on a school senior leadership team. Trained as a British Schools Overseas inspector. |
| p6 | MSc in Management (FinTech & Digital Banking): taught modules complete with an 84% average, dissertation pending. PGCE. BSc Pharmaceutical Science. |
| p7 | Builds with AI coding tools, including Claude Code. Not a trained software engineer. |
| p8 | Served in the RAF from 2002 to 2007 as an electronic technician on radar systems, holding SC clearance. |
Host acquisition agent pipeline
project page →| h1 | An orchestrated pipeline of 11 working agents and rule-based gates. Two further agents are registered but not yet built out, and are not counted. |
| h2 | It worked through all 117 property managers on the Costa del Sol: finding decision-makers, researching each company and drafting personalised outreach. |
| h3 | Around 20 managers agreed to list about 1,000 properties in under three weeks. These are agreements to list, not live listings. |
| h4 | Email outreach got about 2 replies. A phase running email and WhatsApp side by side, each manager on one channel, then a full switch to WhatsApp got about 45 replies (between 40 and 50). |
| h5 | Models are tiered by task: Claude Haiku to sort, Sonnet to research, Opus to write. |
| h6 | Every draft passes an automated tone check (rules, then a model judgement). Failed drafts are rewritten up to three times. |
| h7 | A rule-based compliance gate checks throttling, GDPR lawful basis and consent before every send, and writes an audit record. |
| h8 | Nothing is sent without a person approving it. The agent that handles replies never replies by itself. |
| h9 | Failure handling: automatic fallback to a smaller model, an alert when fallback passes 10% in an hour, retries with backoff, a daily spend cap and kill switches by agent, channel and region. |
| h10 | Several agents use no model at all (prospect finding, sending, compliance, follow-up templates). That is deliberate. |
AI job matching engine
project page →| j1 | A five-model pipeline that routes each task to the cheapest model able to do it well: a low-cost gate model (Jev), Claude Haiku for extraction, Sonnet for matching, Opus for CV parsing and writing, with DeepSeek as a second provider. |
| j2 | Rules and the gate settle 82% of job assessments. The mid-tier model reads 18%. |
| j3 | The gate rejects 47% of candidates in under half a second. |
| j4 | Median AI cost is $0.09 per search run. |
| j5 | Failover to DeepSeek is automatic when the primary provider errors, times out or rate-limits, after three retries. It never triggers on a bad request or a refusal. Anything DeepSeek scores is re-scored by the primary model on the next run. |
| j6 | Each user has a hard monthly AI budget. When it is reached, matching continues on rule scores alone. |
| j7 | Live with paying users since September 2026. 74 of 78 production runs succeeded; all four failures were in the launch cut-over. |
| j8 | The job data feed costs more than the AI. |
Decision Intelligence
project page →| d1 | A guest-facing feature that scores each home as a percentage match against the guest's priorities, using measured distances, noise, sunlight, listing details and reviews. |
| d2 | Runs only when the guest asks. It never books and never reorders results. |
| d3 | AI judgement is labelled separately from measured fact. |
| d4 | Multi-model design: Jev, TypeSafe's decision model, makes the scoring decisions, and a language model reads the guest's request and explains the result. In production within weeks of Jev's public launch. |
| d5 | Shipped in stages: a hand-written evaluation set with a 100% pass bar, then a two-day shadow run against a standard LLM pipeline, then a two-day canary over several hundred live requests. |
| d6 | The launch gate was zero false claims: never calling a home close to the beach when it is not. |
| d7 | The model's instructions were iterated against the evaluation set to fix place ambiguity, such as an Irish bar matching an Irish cultural centre. |
| d8 | If a model fails, it falls back to measured facts alone. This kept the feature working through a model outage in September 2026. |
| d9 | Guest outcome data (usage, bookings) had not been collected by handover in October 2026. |
AI Help and host assistant
project page →| a1 | A guest help assistant grounded in a 497-entry knowledge base. For each question it retrieves the closest entries and answers from them. |
| a2 | It is barred from stating numbers, policies, dates or features that are not in its sources, and points unresolved issues to support. |
| a3 | This was built after the assistant was found inventing facts such as dates. |
| a4 | A separate host assistant drafts profile and listing text and can change only a fixed set of fields. |
| a5 | The safeguards on AI Help are instructions and grounding. There is no automated checker on its answers. |
Wider product and payments work
on the home page →| w1 | Took AtlasOra from zero to public launch in September 2026: a 198-feature requirements document and 98 features shipped with a CTO and four engineers. |
| w2 | In-person discovery with property managers across Malaga and Marbella. |
| w3 | Designed and specified trigger-released escrow contracts on Base. Guest funds were held in EURC and released at check-in, or automatically 24 hours later, with dispute freeze and admin settlement. It ran in production and is currently switched off for cost reasons. |
| w4 | Automatic cross-border host payouts by Revolut bank transfer once the on-chain release was recorded, so hosts never touched crypto. |
| w5 | 16 live API integrations with property management systems (a 17th built and ready), plus a payment provider and KYC. |
| w6 | Led a 13-person team and hired and developed three product designers. |
| w7 | Shipped on four model providers: Anthropic, OpenAI, DeepSeek and Jev. |
Qubic
on the home page →| q1 | Worked directly with the project's AI scientists, David Vivancos and Dr Jose Sanchez. |
| q2 | Hosted their public technical interviews and supported the launch of their research paper on Aigarth. |
Scope
2 entries| s1 | Shapes model behaviour through instructions, grounding, evaluation sets and graders. Has not fine-tuned model weights. |
| s2 | Works in English. |