Admin panels and ops dashboards built right.
Custom internal software your team actually wants to use. Operations dashboards, support consoles, content-moderation queues, ops-on-call surfaces.
- Starting at
- $20k
- Structure
- Phase-by-phase quote, fixed bid
What you get for the work.
- 01
Built against your real data
We connect to the database, API, or warehouse you already have. No middleware layer, no proprietary connectors.
- 02
Role-based from day one
Roles, permissions, audit log. Not a single admin password shared in a shared 1Password vault.
- 03
Designed for speed of use
Keyboard navigation, bulk operations, undo. Internal tools live or die by how fast a power user gets through a queue.
- 04
Owned by you
Code, hosting, secrets all in your accounts. No vendor lock to a low-code platform that bills per seat.
What we build it on.
- Nuxt 3 + Tailwind
- Or Next.js. Whichever fits the team.
- Drizzle / Prisma + Postgres
- When we own the schema.
- Read-only connectors to your DB / API
- When we don't.
- NextAuth / Clerk
- SSO with Google Workspace or Okta typical.
- Inngest / Trigger.dev
- Background jobs for bulk operations.
- Sentry, Better Stack
- When the tool is mission-critical for ops.
$20k buys this. Bigger scope scales it.
Included at the floor
- One internal tool with up to 6 main views
- Role-based access control with audit log
- Connection to one data source (your DB or API)
- Bulk operations with undo on at least one workflow
- Deployed to your infra or Vercel
- Two-week post-launch support
What scales the number
- Multiple connected data sources with cross-source queries
- Real-time updates (sockets, presence, live queues)
- Complex workflow engines (approval chains, escalations)
- Embedded analytics or BI panels
- Customer-facing read-only mirror (status pages, partner portals)
What a starter internal tool looks like
Six views, role-based access, audit log, one data source, bulk ops with undo, deployed. Six to ten weeks. Built so the ops team feels faster, not slower.
Where the build scales
Multi-source queries are the most common scaler. Workflow engines (approvals, escalations, SLA timers) are their own chapter. Real-time queues with multiple ops users on the same record need careful state design.
What we will not build
Tools nobody on the team wants to use. Tools that duplicate work the database admin could do faster with SQL. Tools whose only feature is a wrapped Airtable view.
Common questions.
- Why custom over Retool / Internal?
- Retool and Internal are great until your team hits the wall of "we cannot ship this specific interaction." Once a tool actually matters, the custom build is cheaper than fighting the platform for a year.
- How long until usable?
- Six to ten weeks at the $20k floor for a tool with six views. The team can use a v1 by week three usually.
- Do you handle SSO?
- Yes. Google Workspace and Okta are the common ones. SSO is in scope at the floor.
- Can we self-host?
- Yes. Docker image, environment variables documented, deploy guide handed off at launch. We've shipped to AWS, GCP, Hetzner, and bare-metal.
Pair this build with.
Build this with us.
Twenty minutes through the intake. A real reply within a work day.