Practical playbooks for building on Paddington
No-fluff recipes. Each one gets you live in production on Paddington in fewer than seven explicit steps — landing behind Org Only access by default, one Promote away from the world.
For founders, engineering leaders, and ops teams deciding what to build — and how to own it.
Build a Salesforce Replacement
For revenue and ops teams tired of per-seat CRM pricing and a platform they can't leave.
A CRM is a database, a workflow engine, an auth model, and a UI. You don't need Salesforce's runtime — you need those primitives, owned. On kilter they're catalog services, and the app on top stays small. Live in production in six steps.
Read the playbook →2 min readBuild a ServiceNow Replacement
For IT and platform teams who want ITSM without the ITSM invoice — or the ITSM rollout.
ServiceNow is a workflow engine wearing a CMDB and a portal. The engine is the product. On kilter, durable workflows plus one canonical record store plus governed access give you incidents, changes, approvals, and SLAs — with a thin app. Live in six steps.
Read the playbook →2 min readReplace Supabase with a Better Solution for Vibers and Engineers
For builders who love the Supabase DX but not its ceiling — vibe-coders and engineers alike.
Supabase is Postgres, auth, storage, and realtime — glued and hosted. Great until you hit the walls: opaque limits, egress bills, no exit. Kilter gives you the same primitives as Kubernetes services you own. Same DX, no ceiling, eject anytime. Live in six steps.
Read the playbook →2 min readWhy Auth and Workflow Are More Important Than Your Code
For anyone deciding where to spend their next month of engineering.
Your code is the cheapest, most replaceable part of your system — an agent can rewrite it in an afternoon. What's load-bearing is who can do what (auth) and what must happen in order, exactly once, auditably (workflow). Put those in infrastructure and your app shrinks to glue.
Read the playbook →2 min readFor engineers and builders shipping real systems on the platform.
How to Build a Better Sovereign Supabase for Your Needs
For engineers who want Supabase's primitives without its ceiling — and want to shape them.
The engineer's build: compose Postgres, Ory, object storage, search, and durable workflows into a Supabase-shaped stack you own and can shape. Row-Level Security, JWT you control, local parity with prod, and a clean eject. Live in seven steps.
Read the playbook →2 min readI've Got the Polyglot Blues: TypeScript Scales in Your Mind, Not Your Infrastructure
For teams whose stack is TypeScript-shaped but whose problems aren't.
TypeScript is a great way to think — but rewriting the search, queue, or ML bit in TS because that's what you deploy ships worse systems. Pick the language that fits the problem; on kilter every language is the same deploy. Live polyglot in seven steps.
Read the playbook →2 min readNeon on Kilter: How the Tiers Map, What It Costs, and What's Coming
For engineers weighing Neon's serverless Postgres against a sovereign, full-stack container model — with the pricing worked out.
A detailed read of Neon's Free/Launch/Scale tiers, mapped onto Kilter containers and the published rate card. Neon branches the data; Kilter branches the environment. Includes a compute-unit cost comparison, a small/medium-app bill on both, the hobbyist's 15-project math (Neon vs Supabase), a full-stack table, and an honest list of what Kilter can't do yet — but is building.
Read the playbook →12 min read