Housing back office

// The problem

A housing co-op keeps its members, units, housing charges and subsidies in spreadsheets, posts charges to its books by hand, and finds board documents in a shared drive nobody can search.

// What we would build

One web system for the co-op office and its members: the member and unit register, housing charges with subsidies applied, pre-authorised debit and card payments, meeting notices and minutes, and board documents with access by role. Charges and payments post to QuickBooks Online through its API. Subsidy rules come from your operating agreement and your funder; we build to them.

// What is in it

// Stack

• Next.js
• PostgreSQL
• Stripe
• QuickBooks Online API
• DigitalOcean Toronto

// The problem

A nonprofit housing provider runs its waitlist, offers and annual income reviews in a Microsoft Access database built years ago by a volunteer who has since left, and only one staff member knows how to run it.

// What we would build

We start with a short review of the old database, the forms behind it and how your staff really use it, and write down what to keep and what to drop. Then we rebuild it as a web system, move every applicant and household across with a check against the old totals, and retire the Access file. Applicants can see their status online in English and French. Priority rules are the ones your board and funder set.

// What is in it

// Stack

• Next.js
• PostgreSQL
• Access export scripts
• AWS Canada

// The problem

Maintenance staff get repair requests by phone and text, inspect units at move-in and move-out on paper, and nobody can see a unit's history when a dispute about damage comes up.

// What we would build

An app for maintenance staff on iPhone and Android, and a web portal for residents and the office. Residents report a repair with photos, the office assigns it, and staff close it with notes and parts used. Move-in and move-out inspections are done room by room on the phone with photos, signed by the resident, and kept against the unit.

// What is in it

// Stack

• React Native
• Next.js
• PostgreSQL
• S3-compatible storage
• DigitalOcean Toronto

// The problem

A strata corporation's AGM votes are counted by volunteers from paper ballots and proxies, and when a special resolution passes by a few votes, owners ask how they can be sure it was counted right.

// What we would build

Online notices, registration, proxies and secret ballots, with each ballot's fingerprint and the final tally recorded on a small private ledger run by the corporation and its property manager, so each owner can check their vote was counted without showing how they voted. A well-run online voting service with an audit log often does this job; the ledger helps when a contested vote needs a record neither the board nor the vendor can change. Your bylaws and the governing act set the rules; your board confirms them.

// What is in it

// Stack

• Next.js
• Hyperledger Besu (private, permissioned)
• PostgreSQL
• DigitalOcean Toronto

// The problem

A social housing provider's residents speak a dozen languages, the office phone rings all day with the same questions about rent, repairs and rules, and the written guides exist only in English.

// What we would build

A help centre with articles on rent, repairs, rules, moving and community programs, in the languages your residents speak, and an assistant that answers from those articles and hands over to staff with the conversation attached. We use a frontier model through its API, because the traffic is modest and the articles are public; resident details stay out of what the model sees.

// What is in it

// Stack

• Next.js
• Claude API
• PostgreSQL with pgvector
• Crowdin
• DigitalOcean Toronto

// The problem

A housing society with many buildings has vacancies in its property system, work orders in another, arrears in its accounting software and reserve fund component data in a consultant's spreadsheet, and the board sees none of it together.

// What we would build

A dashboard for the board and senior staff that pulls vacancy, arrears, work orders and reserve fund component data into one small warehouse every night, with the building, the trend and the source behind every figure. We run it on managed cloud hosting with monitoring and backups. The reserve fund figures are your consultant's; we present them, we do not certify them.

// What is in it

// Stack

• Next.js
• PostgreSQL
• dbt
• Property and accounting system APIs
• AWS Canada