Otto · Giving homes memory, history and intelligence.
I founded, designed and built Otto, an AI-native home-ownership platform that turns inspection reports, documents, photos and maintenance information into a persistent home profile and prioritized plans.
I owned the product end to end, from research, product strategy, information architecture and interaction design through the design system, production frontend and backend, data model and AI inference, using modern coding tools and agents.
Empowering Homeowners
Home information is scattered across reports, inboxes, photo libraries, and binders that are easy to lose and hard to trust. When something needs attention, homeowners have to reconstruct context from memory instead of from a single record. Inspectors, contractors, and insurers each keep their own version of the same house, and none of them stay in sync. The result is delayed decisions, duplicated work, and a home that never quite becomes knowable.
Our careers, health, finances, and even our cars have persistent digital records. Our homes do not. Information about a home remains scattered across inspection reports, invoices, warranties, photographs, email threads, and individual memory. I founded Otto to give every home a living digital profile: a structured record of what it is, what has happened to it, and what should happen next.
Research
I conducted more than two dozen interviews with homeowners, friends, family members and past clients. I aggregated the conversations in NotebookLM and several patterns were striking. Most homeowners described managing their home as a meaningful source of stress and anxiety. Despite how valuable their homes were, both financially and emotionally, much of their information remained scattered across paper records, emails, text messages, spreadsheets and individual memory.
The Most Common Concerns of Owning a Home
- Anxiety about unseen and aging systems
- Uncertainty about what a home needs
- Reactive, unpredictable expenses
- Friction over shared responsibilities
“Nobody has taught you anything about how to maintain property... you lack all the things you need to understand to not get swindled and to not feel stupid. And you can't know them all.”
“We're already kind of playing chicken with the roof. Like that is, when I think about it, causes me immense stress.”
“This house is the most stressful part of our life.”
The Home Intelligence Loop
Otto turns home information into structured understanding, action and persistent history.
An inspection produces a finding, recommendation, task, plan, completed work, and home history. The loop then appears as document, Ask Otto, recommendation, task, plan, completed work, and home history.
The Core Loop
Otto’s core loop turns information about a home into action. Documents are uploaded, interpreted, and transformed into clear priorities and next steps. As work is completed, that history flows back into the home’s record, making Otto more useful and intelligent over time.
Create an Otto account
Access to Otto starts with an invite code. Invitees enter the code from the landing page, then create an account using OAuth or an email magic link.
Create new property
A new account starts empty: add a property, search its address, and give it a nickname. You then choose a type, skip or enter purchase details, and confirm the record before Otto creates the home.
Upload your home inspection
Once a property is created, users are encouraged to upload a home inspection or home documents. Otto parses a home inspection and uses the findings to populate the home profile’s nine systems. It also prioritizes the home inspection findings by severity, and orders them relative to when each finding should be attended to.
Otto fills out your home profile
After parsing, Otto fills the home profile with nine systems and their open findings. Opening a system shows each finding in context, then the prioritized tasks that should follow.
Ask Otto
Ask Otto is an integrated LLM utilizing OpenAI's GPT-5.6 Terra model. Homeowners can ask any question regarding their home's documents, history, and data and get clear, useful answers. It helps homeowners understand what matters, what to do next, and why.
As homeowners upload more documents about their home, Ask Otto learns more about its systems, documents, repairs and history. Otto's answers become increasingly accurate and contextual. Over time, Ask Otto evolves from a simple chatbot to having a more complete understanding of a homeowner's home.
Homes are multiplayer
Making the home’s implicit social graph explicit
Homes do not exist in a vacuum. They are inherently multiplayer and have their own implicit social graph. Otto makes that social graph explicit.
Otto makes that network explicit by placing the home, rather than the individual user, at the centre of the product. People join through defined roles, while documents, findings, tasks, conversations, and completed work remain attached to the property itself.
The result is a shared, persistent record of the home that can grow with the people who own, maintain, and care for it.
This led to several foundational product decisions:
1. Membership and roles
A person’s relationship to a property is represented explicitly, allowing for owners, co-owners, family members, tenants, realtors, inspectors, contractors, and other participants.
2. Scoped access
Not every participant should see or change everything. Access can reflect the person’s role and the work they are involved in.
3. Multi-property support
The same person may participate in several homes, and may have a different role in each one.
The first version of Otto prioritizes owners and co-owners, but I designed the underlying model to support the broader network that naturally forms around a property.
This changed Otto from a personal home-maintenance dashboard into something more durable: a shared system of record through which the people responsible for a home can coordinate decisions, contribute knowledge, and preserve its history over time.
Building Otto
From design tokens to production software
I built the Otto Design System from primitives such as colours, typography and spacing through to complex product shells, custom workflows and highly specific, object-level components.
Object model development
Otto’s object model was developed to give the home a persistent structure beneath the interface, connecting documents, findings, tasks, plans, equipment, people, and history around the property itself. Each object was defined across nine dimensions, including its properties, origin, relationships, state, behaviour, status, permissions, lifecycle, and history. Defining these relationships early helped turn Otto from a collection of features into a coherent system that could grow over time without losing its underlying logic.
Property-centred architecture
Information belongs to the home rather than being trapped inside one person’s account.
Domain canonicity
The product’s main objects exist independently and users participate in them.
Identity canonicity
Most objects are understood through their relationship to a user.
Typical identity-centric social product
identity-centric
Most homeowner software treats the home as data inside an account.
Otto treats the home as the durable object and people as participants in its history. This makes the property transferable: ownership, access, and roles can change over time without fragmenting or resetting the home’s underlying record.
The broader idea is that a home deserves a persistent profile of its own, one that accumulates documents, maintenance, improvements, warranties, and institutional memory across its entire life.
Otto
property-centric
The Production Stack
- LinearProject management
- Next.jsReact web framework
- Shadcn UIComponent library
- SupabaseDatabase
- VercelWeb hosting
- Codex & CursorCoding agents
- ResendEmail API platform
- TurbopufferServerless vector search
