Posts on Erkid Hoxholli

10 Practical Tips for an Engineering-Grade AI SDLC

Thu, 08 Oct 2026 00:00:00 +0000

A Note on “Vibe Coding”:
There is nothing wrong with “vibe coding” in the right context. If you are building a quick prototype for a hackathon, trying a new library, experimenting with UI design without a database, or making a small internal tool where bugs do not cause real damage, asking an AI to quickly generate code is totally fine.

This post is not for that.
If you are building production software—especially where users trust you with personal data (PII), health information (PHI), payment details, or mission-critical workflows—users deserve software that is secure, stable, and well-tested. You cannot rely on luck when real people and paying customers depend on your systems.

Using Temporal to Optimize LLM Requests

Wed, 23 Sep 2026 00:00:00 +0000

When building AI-powered features, two operational concerns appear immediately: latency and cost. Calling an LLM third party service typically takes 1 to 3 seconds and incurs token costs on every request.

We recently tackled this while building AI features for Molita.co. Imagine 10 users asking the exact same question (for example, “What is Ikigai?”) at the same time. In a naive setup, your backend fires 10 simultaneous API calls, burns 10x the tokens, and risks hitting external rate limits (429 Too Many Requests).

Preparing a Codebase for Horizontal Scalability (On-Premises / Self-Hosted)

Sat, 12 Sep 2026 00:00:00 +0000

Many early-stage startups and solo developers begin with a classic, streamlined stack: a React SPA frontend, a single Node.js REST API, a single database instance, and Caddy acting as a reverse proxy and SSL terminator.

A Note on Cloud vs. On-Premises:
If you run on managed hyperscalers (AWS, GCP, Azure), much of horizontal scaling is abstracted away behind managed services: AWS Application Load Balancers (ALB), CloudFront, ElastiCache, and Amazon RDS with built-in connection proxies and read replicas.