ShieldThemes Web Development
+1 (415) 555-0142 Get a quote →
THE JOURNAL

Notes on building, shipping & securing the web.

How we shipped a support agent that resolves 62% of tickets
FEATURED · AI
How we shipped a support agent that resolves 62% of tickets
The architecture, evals and guardrails behind a production AI agent — and the three mistakes we made first.
Daniel Reyes
Daniel Reyes
Sep 18, 2026 · 5 min read
All AI WordPress Shopify Performance Security SEO E-commerce Design DevOps Engineering
RAG for small teams: a knowledge base in two weeks
AI · 5 min
RAG for small teams: a knowledge base in two weeks
A pragmatic stack for turning your docs into an assistant your team actually trusts.
Daniel Reyes · Jul 28, 2026
Fine-tuning versus better prompting: a practical decision guide
AI · 5 min
Fine-tuning versus better prompting: a practical decision guide
When a fine-tuned model is worth the data work and maintenance, and when prompting, retrieval and a good eval set get you there faster.
Daniel Reyes · May 12, 2026
Building an AI sales assistant that reps keep using
AI · 5 min
Building an AI sales assistant that reps keep using
Why most sales copilots are abandoned within a month, and the design choices that made one stick with a 40-person sales team.
Daniel Reyes · Feb 24, 2026
Voice agents on the phone line: latency, handoffs and honesty
AI · 5 min
Voice agents on the phone line: latency, handoffs and honesty
What it takes to put an AI voice agent on a real phone line, from sub-second response times to clean transfers and clear disclosure.
Daniel Reyes · Dec 09, 2025
Giving AI agents tool access without handing over the keys
AI · 5 min
Giving AI agents tool access without handing over the keys
Scoped credentials, server-side policy, confirmation steps and prompt-injection defenses for agents that can take real actions in your systems.
Daniel Reyes · Oct 15, 2025
Keeping LLM costs predictable as usage grows
AI · 5 min
Keeping LLM costs predictable as usage grows
Budgets, caching, model routing and monitoring that stop a successful AI feature from turning into an unpleasant invoice.
Leo Tanaka · Aug 06, 2025
When workflow automation beats an AI agent, and when it does not
AI · 5 min
When workflow automation beats an AI agent, and when it does not
A decision framework for choosing between deterministic automation, a single LLM step and a full agent, with cost and reliability trade-offs.
Daniel Reyes · May 27, 2025
Automating invoice intake with LLMs without losing the audit trail
AI · 5 min
Automating invoice intake with LLMs without losing the audit trail
A document-processing pipeline that extracts invoice data reliably, flags what it is unsure about, and keeps finance and auditors comfortable.
Daniel Reyes · Mar 19, 2025
Write the evals before the prompt: testing LLM features properly
AI · 5 min
Write the evals before the prompt: testing LLM features properly
How we build evaluation sets, graders and CI gates for LLM features so that every prompt or model change is measured instead of guessed.
Daniel Reyes · Jan 14, 2025
One email a month. Only the good stuff.
Field notes from real client projects. No spam.