Engineering blog
Deep-dives, case studies, and lessons learned from shipping production systems across AI, fintech, healthcare, and more.
Why Open-Source Maintainers Are Right to Reject 90% of AI Pull Requests: The Rise of Invariant Engineering
An institutional systems investigation into the crisis of synthetic slop in open-source review queues, why traditional LLM wrappers fail at the compiler boundary, and how deterministic invariant verification restores maintainer trust.
Open-Source Maintainers Are Right: Why I Respect the Pushback Against Low-Effort AI PRs
Founder Memo by Patrick Ribbsaeter: Why open-source maintainers closing unvetted AI pull requests is rational self-preservation, what physical GPU hardware diagnostics taught me about software invariants, and the upstream receipts behind ZeroSlop™.
The KV Cache Memory Wall: How Multi-Gigabyte Context Windows Throttle GPU Concurrency on H100 & B200 Clusters
A frontier AI infrastructure investigation into the mathematical mechanics of Key-Value (KV) cache HBM exhaustion, the asymmetric conflict between compute-bound prefill and memory-bound decode, and Disaggregated Prefill-Decode (PD) architectures.
The Stealth Attack on Enterprise CI: Why Autonomous Coding Agents Leak Developer Paths and API Secrets
An institutional systems investigation: why autonomous coding assistants routinely leak developer environment paths and active credentials into public PRs, and why deterministic pre-flight airlocks are replacing late-stage CI scanners.
The Rate-Limit Trap: Why AI Diff Sprawl Is Freezing Enterprise Merge Queues
Founder Memo: When pull requests expand from 20 focused lines to 400 lines of AI boilerplate, delivery velocity collapses. True systems engineering is having the discipline to isolate root invariants and eliminate code review drag before human review.
The Economics of Diff Minimization: Why 500-Line Pull Requests Are Balance Sheet Liabilities
An institutional analysis of code review microeconomics: why AI-generated diff sprawl destroys engineering velocity, degrades review rigor, and how deterministic diff minimization cuts review latency by 50% to 70%.
Seniority Is Diff Minimization: What 20 Years in Systems Engineering Taught Me About AI PRs
Founder Memo: Junior engineers measure productivity in lines added. Senior systems engineers measure mastery in lines preserved. How autonomous coding assistants inverted this discipline—and how to reclaim it.
The Failure of Probabilistic AI Code Review: Why Enterprise CI Demands Compiler-Grade Invariants
An engineering investigation into why LLM-based PR reviewers hallucinate bugs, miss critical security flaws, and create reviewer fatigue—and why deterministic invariant airlocks are replacing them.
Why I Refused to Build Another LLM Code Reviewer: A Founder's Notes on Software Invariants
Founder Memo: When the industry rushed to build chatbot wrappers that leave 30-paragraph comments on pull requests, I walked away. Here is why we built a deterministic pre-flight airlock instead.
Ribbsaeter Systems Releases ZeroSlop™ on GitHub Marketplace: The End-to-End PR Invariant & 12-Shield Security Airlock
Official Press Release: Ribbsaeter Systems launches ZeroSlop™ on the GitHub Marketplace—an end-to-end pre-flight pull request invariant and 12-Shield CWE security airlock engineered to strip 94% of AI slop and get PRs merged 100% green in minutes.
Synthetic Supermodels in Production: How AI is Transforming Luxury Fashion, Commercial Casting, and Brand IP
An architectural and strategic deep-dive into synthetic model casting: character facial consistency, 10-locale headless delivery, perpetual likeness licensing, and how luxury houses are eliminating physical production bottlenecks.
Building Custom MCP Servers for Enterprise AI: A Practical Guide
How Model Context Protocol (MCP) servers expose internal tools and data to LLM agents in production — architecture, authentication, security and the mistakes we've stopped making.
TypeScript Monorepos at Scale: Turborepo, pnpm and the Patterns That Actually Work
A field guide to structuring TypeScript monorepos for teams shipping multiple packages, apps and services — based on what we run in production across fintech, marketplace and SaaS clients.
Building Compliant Healthcare Software in Europe: MDR, GDPR and the Engineering Reality
A senior engineering team's guide to shipping healthcare platforms that satisfy EU MDR, GDPR, and national health-data regulations — without drowning in paperwork or sacrificing delivery speed.
React Server Components in Production: Six Months of Real-World Lessons
What we learned migrating a high-traffic e-commerce platform to the Next.js App Router with React Server Components, streaming, partial pre-rendering and Server Actions.
The Next.js Performance Audit We Run on Every Client Project
Our 30-point performance checklist for Next.js applications — covering Core Web Vitals, bundle size, server components, image optimization and the database queries nobody thinks to profile.
Fintech Compliance Checklist for 2026: PSD2, PCI-DSS 4.0, GDPR, KYC and AML
A practical engineering checklist for building compliant fintech products in 2026 — from PSD2 strong customer authentication and PCI-DSS 4.0 to GDPR data minimisation and Travel Rule for crypto.
Marketplace MVP Architecture: The Stack and Decisions That Get You to Launch
How we architect two-sided marketplace MVPs — from payment splits and trust systems to the database schema decisions that determine whether you can scale past launch.
Scaling PostgreSQL with pgvector for RAG: Index Strategy, Partitioning and HNSW Tuning
Index strategies, partitioning patterns, and HNSW tuning for retrieval-augmented generation systems serving millions of vectors at sub-100 ms p95 latency on commodity Postgres.
Docker in Production: The Practices That Survive Real Incidents
Production-proven Docker practices for production workloads — multi-stage builds, security hardening, health checks, log management and the Compose-to-Kubernetes migration path.
Migrating to React Native's New Architecture: Fabric, TurboModules and the Bridgeless Future
A pragmatic playbook for adopting Fabric and TurboModules in existing apps — from preflight audits, codegen, native module migration, to gradual rollout in production.
MongoDB vs PostgreSQL in 2026: When to Use Which and Why We Default to Postgres
A practical comparison of MongoDB and PostgreSQL for production applications — covering data modelling, transactions, scaling, developer experience and the specific use cases where each database wins.
Design Systems as Product, Not Project: How to Stop Accumulating UI Debt
Why treating your design system as an internal product — with versioning, changelogs, a roadmap and a real owner — dramatically reduces UI debt and accelerates every feature team that consumes it.
Engineering Real Estate Platforms: Listings, Maps, Lead Routing and the Data Challenges Nobody Warns You About
A technical deep-dive into building real estate platforms — covering listing data pipelines, map integration, lead routing, IDX/MLS compliance and the performance patterns that make property search feel instant.
Migrating from Laravel to Next.js: A Practical Guide for Growing Teams
When and how to migrate a Laravel monolith to Next.js — covering the strangler-fig pattern, API extraction, authentication migration, database considerations and the cases where staying on Laravel is the right call.
Cutting Kubernetes Costs by 60% Without Sacrificing Reliability
A real case study in right-sizing nodes, pod-disruption budgets, spot instances, Karpenter-based autoscaling, and the FinOps practices that turned a $48k/month cluster into a $19k one.
Logistics Software Architecture: Real-Time Tracking, Route Optimisation and the Integration Layer That Makes or Breaks the System
How we architect logistics and supply-chain platforms — covering real-time vehicle tracking, route optimisation, warehouse management integration, and the event-driven patterns that keep everything in sync.
The State of Headless Commerce in 2026: Shopify Hydrogen vs. Medusa vs. Saleor vs. BigCommerce
A practical comparison of the four headless commerce backends we ship with most often — strengths, weaknesses, real total cost, and how to choose for your stage and scale.
Want our insights in your inbox?
Occasional engineering deep-dives, no fluff. We respect your time.
Start a project intake