
Prisma in production: great developer experience, hidden costs, and where raw SQL wins
Where Prisma's schema-first workflow pays off in production, which costs it moves onto your team, and the query shapes that still belong in SQL.
What we've learned shipping MVPs, choosing architecture, and running production software for founders - written up as we go.

Where Prisma's schema-first workflow pays off in production, which costs it moves onto your team, and the query shapes that still belong in SQL.

Transactional enqueue, idempotency, retry strategy and dead-letter handling - plus database queues, Redis, RabbitMQ, Kafka, SQS and workflow engines compared.

How B-trees work, why composite column order decides everything, and the twelve specific reasons a perfectly good index gets ignored by the query planner.

Lock queues, expand and contract, safe recipes for every common ALTER TABLE, batched backfills, and the CI checks that stop unsafe migrations reaching prod.

A working guide to Docker - how containers function, Dockerfiles that build fast, Compose, security defaults worth changing, and the downsides nobody mentions.

A detailed guide to Git - what it actually stores, the everyday commands, rebasing, undoing mistakes, bisect, large repos, and what Git 3.0 changes.

Schema design, resolvers, N+1, caching, federation, security and complexity limits - an honest account of what GraphQL costs and when it pays for itself.

Protocol Buffers, streaming, service contracts, browser limits, observability and versioning - what gRPC costs, and when plain HTTP APIs are enough.

A working guide to nginx - the process model, location matching, reverse proxy and cache patterns, rate limiting, debugging, and what changed this year.

A working guide to PostgreSQL 18 - JSONB, indexes, EXPLAIN, queues, full-text and vector search, production operations, and a learning path that works.

A working guide to modern Python - uv, type hints, asyncio, free-threading, performance, production tooling, and an honest look at where it falls short.

A working guide to Redis - data structures, expiry and eviction, persistence, caching and locking patterns, security, and the Valkey licensing split.

How to design REST APIs you can still change in three years - breaking-change rules, cursor pagination, RFC 9457 errors, and contracts enforced in CI.

What each isolation level really prevents, why Repeatable Read still allows write skew, how deadlocks happen, and how to write a retry loop that works.

Connection lifecycle, authentication, reconnects, presence, horizontal scaling with Redis, backpressure and failure handling - the parts demos leave out.

A practical look at Bun 1.3.14 - what the all-in-one JavaScript runtime does well, where it still breaks, and how to decide if it belongs in production.

A working guide to Hono 4.13 - Web Standards routing, typed validation, the RPC client, deployment across runtimes, and where the framework falls short.

A working guide to Next.js 16 - Cache Components, Turbopack, Server Functions, self-hosting, the security cadence, and what the upgrade actually costs.

How QUIC works, why HTTP/3 wins on mobile and loses on fast fixed lines, how to deploy it on nginx or a CDN, and what to measure afterwards.

Why Rust keeps replacing C, C++ and JavaScript tooling - the safety and speed case, the real costs, and how to adopt it without betting the company.
Tell us what you're building and we'll come back with scope, a timeline and a price.