Search
Start AI Tools with Intent Solutions
Daily work journal below. Real case studies, deep-dives, and operator-grade write-ups on AI engineering, DevOps, and the messy reality of shipping software that touches LLMs, by Jeremy Longshore at Intent Solutions.
Curated long reads
Let the Model Judge. Make the Code Decide.
The model owns judgment. Deterministic code owns the gates, the commit, and the audit trail. How I rebuilt a daily writing pipeline so a hung session can no longer brick tomorrow.
Adversarial Review: The Six Lenses That Halted a Rollout
Six lenses, live state, eighteen risks. I thought the safety work was shipped. The review proved three assumptions were theater. The durable rule: the model proposes; the system owns identity, trust, and state.
The Moat Is the Trust Layer
A local-RAG demo became Intent NEXUS. The moat is not the chatbot. It is egress control, code-enforced citations, a ledger you can check, injection defense, and evals that cannot lie quietly to you.
Recent posts
The Third State: When Your Checkout, Image, and Docker Volume All Disagree
A rebuild succeeded and nothing changed. How Docker named volumes shadow your image, and how to write a deploy smoke check that asserts on the served bytes.
ReadNow-LMS 2.0 and the Email Cutover
A 2.0.0 that is mostly a careful rename plus security headers is exactly what a healthy release looks like.
ReadSplitting Privileges at the CI Boundary
Fork CI privilege separation in GitHub Actions prevents secret exfiltration. Run untrusted fork code and trusted deterministic steps safely.
ReadGood mechanisms are not an architecture until a doctrine names them
You can ship a dozen good mechanisms and still have no architecture. A doctrine names why they cohere and forces every capability to fail closed.
ReadWrong-Mode Green Is Not a Gate
A freshness gate that accepts the wrong mode flag and still exits green trains people to trust a board that cannot see. Pin the argv semantics, not just the script path.
ReadTemporary Is Not a Plan: Fork Discipline for an Adopted LMS
Adopting an open source LMS got us live in days. What keeps a fork from becoming a private product: a named upstream path and a written retirement condition.
ReadDo Not Blindly Restart: Designing a Self-Healing Watchdog That Stays Honest
A burn-in watchdog latched on a phantom io spike. The fix: fail-closed self-healing that restarts an allowlist, latches everything else, and never trusts one sample.
ReadSeries & curriculum
MCP for Beginners
End-to-end Model Context Protocol curriculum — solutions in Python, TypeScript, Java, Rust, C#, .NET — translated across six languages.
Agentic Design Patterns
Patterns and anti-patterns for production agent systems. Decision frameworks, prompt scaffolds, evaluation harnesses.
Tiny Recursive Models
Building small recursive systems where simple loops compose into emergent behavior. Series in progress.
Research & Curriculum
Long-form research articles, learning paths, and reading lists for the AI builder community.
Retrospectives
June 2026: The First Full Month Under the Repaired Rubric — Tier 1 Went From 5% to 50%, and the Last Mile Got Louder
June 2026 retrospective: the first full month running under May's repaired tier rubric. The distribution corrected from 5% to 50% Tier 1, exactly as predicted — but the structural sweep still flags every disagreement as inflation, and no human has adjudicated one. 27 posts, 1,166 commits.
May 2026May 2026: The Month the Classifier Caught Itself — A Mid-Month Calibration Reckoning, 27 Posts, and 964 Commits
May 2026 retrospective: the tier classifier caught itself inflating. Calibration found the rubric structurally broken mid-month. 27 posts, 964 commits.
Let's build something that ships to production
Intent Solutions designs, builds, and ships production AI systems: Claude, MCP, RAG, agents, and the boring infrastructure that makes them trustworthy. Tired of AI demos that never make it past the POC?
No commitment required. We respond within 24 hours.