Writing

A practical guide to SBOM diff gates for AI coding runners using lockfile-linked SBOM snapshots, severity-aware policy checks, provenance verification, and reviewer evidence so dependency-changing agent patches stay fast without turning into supply-chain blind spots.

8 min · AI Coding · Supply Chain · SBOM

A practical guide to enforcing network egress policies for AI coding runners with allowlisted registries, DNS-aware controls, staged bootstrap lanes, and audit logs so agents cannot quietly exfiltrate data while still completing real builds and tests.

6 min · AI Coding · Runner Security · Network Policy

A practical guide to defining SLOs and burn-rate alerts for AI coding gateways with first-token latency, tool success budgets, lane-aware error budgets, and fallback policy so operators catch drift before slow turns become broken automation.

6 min · AI Coding · SLOs · Gateway Ops

A practical guide to enforcing path-scoped write permissions for AI coding agents with policy maps, CODEOWNERS alignment, pre-apply checks, and reviewer evidence so one helpful patch cannot quietly edit the whole repository.

4 min · AI Agents · Repository Security · CODEOWNERS

A practical guide to catching MCP schema drift with fixture snapshots, contract hashes, compatibility lanes, and CI alerts before tool changes turn into silent agent regressions.

7 min · MCP · Tooling · Reliability

How to defend shared container caches for AI coding runners with digest pinning, tenant-scoped keys, provenance attestations, and restore-time verification.

5 min · AI coding agents · container security · build cache

A practical guide to secret scanning gates for AI-generated patches using diff-aware detectors, entropy filters, allowlists, push protection mirrors, and reviewer evidence so coding agents do not turn speed into credential leaks.

4 min · Secret Scanning · AI Coding Agents · Security Guardrails

A practical guide to prompt cache segmentation for shared AI coding gateways using tenant-aware keys, prefix freezing, trust lanes, TTL classes, and invalidation ledgers so teams get cache wins without leaking context or serving stale prompts.

5 min · Prompt Caching · Gateway Architecture · Multi-Tenant AI

A practical guide to GPU memory headroom policies for local coding model gateways using admission control, KV cache reserves, prompt budgets, and fallback lanes so coding agents stop crashing into OOM roulette.

5 min · Local LLMs · Inference Ops · AI Coding Agents

A practical guide to calibrating LLM judges for AI coding agent evals with anchor cases, pairwise scoring, disagreement audits, and human spot checks so quality scores stay useful instead of turning into prompt theater.

7 min · Evals · AI Coding Agents · LLM Judge

Keep long-running AI coding agents sharp with transcript compaction ledgers, pinned evidence blocks, refresh triggers, and context budgets that cut prompt bloat without losing critical state.

6 min · Context Engineering · Agent Reliability · Prompt Budgeting

How to design tool timeout budgets for AI coding agents with layered deadlines, cancellation hooks, retries, and fallback lanes so hung tools stop burning whole runs.

5 min · Agent Reliability · Tooling · Automation

A practical guide to lease-based workspace locks for parallel AI coding agents using TTL-backed claims, path scopes, heartbeats, and stale-lock recovery so concurrent runs stop clobbering the same files.

7 min · AI Coding Agents · Concurrency Control · Repo Automation

A practical guide to warming pnpm, uv, Docker BuildKit, and CI caches so AI coding agents stop wasting runs on repeated dependency installs and flaky bootstrap latency.

7 min · AI Coding Agents · Build Systems · CI Reliability

Use hermetic wrapper scripts, environment pinsets, path allowlists, and normalized tool output so AI coding agents call CLIs reproducibly across laptops, CI, and remote sandboxes.

5 min · AI Coding Agents · Tooling · Reproducibility

Use checkpointed workspaces, resumable manifests, and verifier-aware rollback points so AI coding agents can recover from bad edits without restarting every long debug loop from scratch.

7 min · AI Coding Agents · Workspace Checkpoints · Recovery

Set artifact retention budgets for AI coding agents so logs, traces, screenshots, and test bundles stay reviewable without turning storage and privacy risk into a quiet tax.

7 min · AI Coding Agents · Artifact Retention · Evidence Ops

How to build diff-scoped test selection for AI coding agents with dependency graphs, coverage fallbacks, and flaky-test guardrails.

5 min · AI Coding Agents · Testing · CI

How to build replayable incident fixtures for AI coding agents with sanitized bug packets, verifier scripts, and deterministic replay loops.

5 min · AI Coding Agents · Incident Response · Debugging

A practical guide to incremental code index refresh for AI coding agents using change feeds, symbol-level dirty sets, bounded backfills, and rollback lanes so retrieval stays fresh without rebuilding the whole repository index.

6 min · Code Retrieval · AI Coding Agents · Indexing

A practical guide to using reranker budgets, exact-match promotion, and latency caps so AI coding agents retrieve the right code paths without turning every lookup into a slow expensive search stack.

6 min · Code Retrieval · AI Coding Agents · Reranking

A practical guide to building browser session fixtures for AI web agents with seeded auth state, DOM waits, replayable traces, and reset hooks so web workflows stay reviewable instead of becoming flaky demos.

5 min · Browser Automation · AI Agents · Playwright

A practical guide to migrating embedding models in production RAG systems with dual indexing, overlap audits, shadow queries, and rollback lanes so better benchmarks do not quietly lower real recall.

6 min · RAG · Retrieval Engineering · Embeddings

How to test MCP tool adapters with golden transcripts, replay harnesses, and schema drift checks so agent tools stay stable in production.

6 min · MCP · Testing · Agent Reliability

How to use Git sparse checkout, cone patterns, verifier lanes, and hydration rules so AI coding agents can work inside giant monorepos without dragging the whole repository into every task.

5 min · Git · Monorepos · AI Coding Agents

How to keep local coding model gateways responsive with warm pools, health probes, memory-aware routing, and queue control for AI coding workloads.

6 min · Local LLMs · Inference Ops · AI Coding Agents

A practical guide to running Git bisect with AI coding agents using verifier scripts, patch checkpoints, and evidence capture so regression hunts stop turning into wide, expensive guessing.

7 min · AI Coding Agents · Git Bisect · Debugging

A practical guide to routing AI-generated patches with CODEOWNERS, risk scoring, and escalation rules so the right humans review the right changes before automation outruns ownership.

6 min · AI Coding Agents · Code Review · CODEOWNERS

A practical guide to detecting repo map drift for AI coding agents with AST snapshots, ownership signals, CI freshness gates, and compact evidence packets so context files stay useful instead of becoming polished fiction.

6 min · AI Agents · Code Retrieval · Repo Maps

A practical guide to approval-aware MCP proxies with policy checks, short-lived claims, replay defense, and audit trails so remote tools stay useful without turning every agent session into ambient access.

6 min · AI Agents · MCP · Security

A practical guide to triaging CI failures for AI coding agents with failure buckets, artifact packs, flaky-test fingerprints, and patch-vs-retry rules so automation stops retrying doomed jobs or patching around broken infrastructure.

7 min · AI Agents · CI/CD · Reliability

A practical guide to building session-aware MCP tools with tenant binding, claim checks, lease scoping, and audit trails so multi-user agent systems stop leaking context or acting in the wrong workspace.

5 min · AI Agents · MCP · Security

A practical guide to debugging RAG recall with overlap heatmaps, chunk boundary probes, citation audits, and reranker spot-checks so retrieval systems stop missing facts that are already in the corpus.

7 min · RAG · Retrieval · AI Search

A practical guide to AI-to-SQL guardrails using read replicas, query budgets, EXPLAIN gates, parameterized templates, and audit logs so natural-language analytics does not turn into runaway database risk.

5 min · AI Agents · SQL Safety · Analytics

A practical guide to reviewing AI-generated frontend changes with Playwright screenshot diffs, trace artifacts, DOM assertions, and evidence packs so UI automation ships with proof instead of hopeful screenshots.

6 min · AI Agents · Playwright · Frontend Testing

Build webhook inboxes for AI agents with signature verification, idempotency keys, durable event ledgers, and retry-safe workers so callbacks stay secure and do useful work exactly once.

6 min · AI Agents · Webhooks · Reliability

A practical guide to securing remote MCP servers with signed tool claims, short-lived capability tokens, audience checks, and replay protection so agents can call powerful tools without carrying ambient access.

7 min · MCP · Security · AI Agents

A practical comparison of OpenClaw, Claude Code, and Cursor for repository automation, covering scheduled runs, approval boundaries, isolation, and review flow so teams can pick the right tool for the right automation lane.

7 min · AI Agents · OpenClaw · Claude Code

A practical guide to reducing tool output for AI agents using schema-aware reducers, taint labels, citations, truncation budgets, and escalation paths so raw CLI and API dumps stop bloating context or hiding the useful signal.

8 min · AI Agents · Tooling · Context Engineering

A practical guide to dry-run lanes for AI agents that execute shell commands using command risk classification, diff previews, ephemeral sandboxes, and promotion gates before anything mutates a real system.

7 min · AI Agents · Shell Safety · Automation

A practical guide to prompt provenance for AI agents using version pins, hashable prompt bundles, behavior diffs, and rollback playbooks so prompt edits stop behaving like invisible production deploys.

6 min · Prompt Engineering · AI Agents · Reliability

Build token spend anomaly detection for always-on AI agents with labels, rolling baselines, budget lanes, and actionable alerts before quiet cost drift becomes a monthly surprise.

6 min · Cost Engineering · AI Agents · Observability

Use policy-as-code gates, Terraform plans, and evidence packs to stop AI-generated infrastructure changes from shipping plausible but unsafe diffs.

6 min · AI Infrastructure · Policy as Code · Terraform

Build ephemeral preview environments for AI-generated pull requests with TTL cleanup, seeded fixtures, smoke tests, and cost guardrails.

5 min · AI Coding Agents · Preview Environments · GitHub Actions

A practical guide to canarying model upgrades for AI coding agents with eval slices, fallback lanes, budget guardrails, and promotion scorecards so model refreshes do not quietly degrade code quality.

6 min · AI Coding Agents · Model Upgrades · Canary Releases

A practical guide to freshness guards for long-running AI coding agents using repo fingerprints, file watchlists, context TTLs, and revalidation gates so stale context does not quietly turn into bad patches.

6 min · AI Coding Agents · Context Engineering · Reliability

A practical guide to versioning MCP servers with capability negotiation, contract lanes, deprecation windows, and compatibility tests so one server upgrade does not break every agent client at once.

6 min · MCP · AI Agents · Compatibility

A practical guide to tool capability manifests for AI agents using side-effect metadata, cost hints, auth lanes, and planner scoring so agents choose safer and cheaper tools on purpose.

6 min · AI Agents · MCP · Tooling

A practical guide to backpressure for AI agents using queue budgets, concurrency caps, lease-aware workers, and overload signals so tool-calling systems stop amplifying latency into self-inflicted outages.

6 min · AI Agents · Concurrency · Queues

A practical guide to using AST codemods, agent-generated transform drafts, fixture-led verification, and rollout checkpoints so large API migrations stay mechanical, reviewable, and much safer than regex-driven bulk edits.

6 min · AI Coding Agents · Codemods · Refactoring

A practical guide to dependency update lanes for AI coding agents using risk tiers, SBOM diffs, focused verification, and staged promotion so automated upgrades stay fast without turning every version bump into reviewer roulette.

6 min · AI Coding Agents · Dependency Management · SBOM

A practical guide to environment manifests for AI coding agents using pinned toolchains, seeded fixtures, and verifier snapshots so patches reproduce the real bug instead of fixing a local fiction.

5 min · AI Coding Agents · Reproducibility · Dev Containers

Design AI agent tool calls to fail loudly, classify errors, retry safely, and surface human-actionable recovery paths instead of burying broken automation behind vague exceptions.

5 min · AI Agents · Tool Calling · Reliability

Build safer AI coding workflows by generating multiple candidate patches in parallel, scoring them with verifiers, and promoting only the most reviewable result.

6 min · AI Coding Agents · Verification · Developer Workflows

A practical comparison of hybrid local and cloud AI development workflows, including routing rules, privacy lanes, fallback policy, and cost tradeoffs for teams that want fast local iteration without giving up stronger remote models when they matter.

7 min · Hybrid AI · Local LLMs · Developer Workflow

A practical guide to secrets management for AI agents using short-lived credentials, brokered access, OIDC, scoped injection, and audit trails so useful automation does not quietly inherit production blast radius.

8 min · AI Security · Secrets Management · OIDC

A practical guide to semantic caching for AI coding agents using fingerprinted prompts, repo-aware invalidation, trust boundaries, and fallback verification so cached answers stay fast without becoming quietly wrong.

9 min · AI Agents · Caching · Developer Tools

A practical guide to verifier pipelines for AI coding agents using fast syntax gates, semantic checks, policy rules, and escalation paths that catch risky edits before PR review.

7 min · AI Agents · Verification · Code Review

A practical guide to capability leases for AI agents, using short-lived access grants, scoped tokens, policy checks, and audit trails instead of permanent tool permissions.

6 min · AI Agents · Security · Access Control

A practical guide to using circuit breakers, budget guards, and cooldown policies for AI agents so model outages, tool flaps, and bad retries do not cascade into user-facing incidents.

7 min · AI Reliability · Circuit Breakers · Agent Ops

How to debug multi-step AI agent failures with trace IDs, span annotations, replay packets, and failure buckets so incidents stop feeling like prompt necromancy.

6 min · Agent Debugging · Observability · OpenTelemetry

How to route coding-agent tasks between fast and expensive models using task scoring, guardrails, verification loops, and fallback policies that improve latency without quietly lowering quality.

5 min · Coding Agents · Model Routing · Cost Engineering

Why huge advertised context windows still break coding agents, and how bounded context packets, rolling summaries, and refresh loops keep long prompts usable.

5 min · Context Engineering · Coding Agents · Local LLMs

A practical guide to rolling out AI features in shadow mode with traffic replay, scorecards, redaction, and promotion gates so teams can test real behavior before enabling user-visible actions.

7 min · Shadow Mode · AI Reliability · Production AI

A practical guide to AI-generated tests, showing where LLM-written unit and integration tests add speed, where they fail quietly, and how to harden them with mutation checks, fixtures, and review gates.

6 min · AI Testing · Software Quality · Pytest

A practical guide to hybrid code retrieval for AI coding agents using embeddings, tree-sitter structure, ripgrep fallback, and reranking so prompts stay smaller and edits stay more accurate.

7 min · AI Agents · Code Retrieval · Tree-sitter

A practical guide to choosing 4-bit, 6-bit, 8-bit, and mixed quantization for coding models on consumer hardware, with concrete runtime tradeoffs for Ollama, llama.cpp, and vLLM.

8 min · Quantization · Local LLMs · Ollama

A practical guide to shipping database changes with AI coding agents using expand-contract migrations, schema guards, dual writes, and rollback checkpoints.

7 min · AI Agents · Database Migrations · Reliability

A practical guide to planner-executor handoffs for AI agents using task manifests, claim checks, verification gates, and rollback-friendly execution so multi-agent workflows stay understandable under real load.

5 min · Multi-Agent Systems · AI Agents · Workflow Design

A practical guide to running scheduled AI workflows with clean task payloads, isolated runs, idempotency guards, and approval-aware delivery so cron automation stays useful instead of becoming a haunted queue.

5 min · Automation · AI Agents · Cron

A practical guide to isolating long-lived AI agents with session-scoped memory, worktrees, tool lanes, and approval boundaries so one task cannot quietly contaminate another.

5 min · Agent Architecture · Session Isolation · Reliability

A practical guide to building eval harnesses for AI coding agents using patch-level tasks, invariant checks, execution sandboxes, and scorecards that catch bad fixes before they reach review.

7 min · AI Evaluation · Coding Agents · Developer Tools

A practical guide to using GitHub Actions OIDC for AI agents so CI jobs can deploy, evaluate, and run ops tasks with short-lived cloud credentials instead of stored secrets.

7 min · GitHub Actions · OIDC · AI Agents

A practical guide to protecting AI agents from prompt injection through tool outputs, with trust boundaries, sanitization patterns, isolation lanes, and reviewable execution guards.

5 min · AI Security · Prompt Injection · MCP

A practical guide to building context packets for engineering agents with repo maps, task manifests, relevance scoring, and bounded evidence bundles that improve edit quality without dumping the whole codebase into context.

6 min · AI Agents · Context Engineering · Developer Workflow

A practical guide to building approval workflows for AI agents that can write code, open PRs, and call tools without turning every useful action into a security incident.

8 min · AI Agents · Approval Workflows · Security

A practical comparison of Ollama, llama.cpp, and vLLM for local AI development, with setup patterns, code examples, performance tradeoffs, and what I would actually run on laptops, workstations, and small servers.

8 min · Local LLMs · Inference · Developer Workflow

A practical guide to building schema-first tool contracts for AI agents with JSON Schema, runtime validation, adapter layers, and fail-closed execution paths that keep tool use reliable and reviewable.

6 min · AI Agents · Tool Calling · JSON Schema

A practical guide to building retryable AI agent jobs with idempotency keys, checkpoints, queues, and human-safe recovery paths.

7 min · AI Agents · Reliability · Queues

A practical guide to choosing MCP transports for local and remote agents, covering stdio, HTTP, and SSE tradeoffs, auth, latency, and deployment patterns.

7 min · MCP · AI Agents · Developer Infrastructure

A detailed step-by-step guide to running Claude Code with Unsloth Qwen3.5-35B-A3B on a Mac mini M4 Pro, including model choice, local inference setup, KV cache fixes, tuning, and practical tradeoffs.

10 min · Claude Code · Unsloth · Qwen

A practical guide to using Git worktrees so Claude Code, Cursor, Codex, and other AI coding agents can work in parallel without stomping on the same checkout, index, ports, or test state.

6 min · Git · Worktrees · AI Coding

A practical guide to using Gemini CLI for terminal-first debugging with plan mode, shell tooling, focused context, and repeatable fix loops that keep incidents moving without turning the model loose on your whole machine.

6 min · Gemini CLI · Debugging · Terminal Workflow

A detailed guide to designing reusable OpenClaw skills with a scheduled repo worker example, including project-memory vs task-engine modes, repo-local state, PR packaging runs, and reviewer-in-the-loop automation.

12 min · OpenClaw · Agent Skills · AI Coding

A practical guide to isolating AI coding agents with dev containers, read-only mounts, network controls, and microVMs so they can work usefully without getting a blank check on your machine.

8 min · AI Coding · Security · Dev Containers

A practical guide to using repo-specific instruction files like AGENTS.md and CLAUDE.md to make AI coding agents safer, more consistent, and easier to review.

8 min · AI Coding · Developer Workflow · Agent Guardrails

A practical guide to using Cursor effectively in large monorepos with layered rules, scoped context, repo maps, verification loops, and guardrails that stop AI edits from wandering across the codebase.

7 min · Cursor · Monorepos · AI Coding

A practical guide to reviewing AI-generated pull requests with scoped diffs, invariant checklists, CI guardrails, and fast human approval loops.

6 min · AI Coding · Code Review · GitHub

A practical guide to reducing latency and cost in tool-heavy AI agents with stable prompt prefixes, prompt caching, context compaction, and retrieval boundaries.

7 min · AI Agents · Prompt Caching · Cost Optimization

A practical guide to tracing AI agents with OpenTelemetry and OpenInference so you can debug tool calls, latency, token usage, and multi-step failures without drowning in logs.

7 min · AI Agents · OpenTelemetry · Observability

A practical guide to building a local LLM development environment with Ollama, Open WebUI, coding tools, quantization choices, prompt/version hygiene, and smart fallbacks when local models are not enough.

8 min · Local LLMs · Ollama · AI Coding

A practical guide to evaluating RAG systems by separating retrieval from generation, building grounded test sets, tracking the right metrics, and running regression checks before users find the bugs.

9 min · RAG · Evaluation · LLM Ops

A practical guide to designing memory for AI agents with working memory, episodic summaries, semantic facts, write policies, retrieval rules, and debugging loops that stay useful over time.

8 min · AI Agents · Memory · Developer Workflow

A practical guide to using Claude Code for large refactors with specs, checkpoints, verification loops, and guardrails that keep big changes understandable.

7 min · Claude Code · Refactoring · AI Coding

A practical guide to hardening OpenClaw on a VPS with least-privilege credentials, network restrictions, approval boundaries, and operations checks that actually matter.

6 min · OpenClaw · VPS · Security

A practical guide to designing MCP servers that stay useful under real agent workloads without turning into a security liability.

9 min · MCP · AI Tools · Security

How I built a collection of 6 browser games using Claude Code and Qwen for AI pair programming.

5 min · AI-Assisted · Games · JavaScript

How I built an AI-powered document-to-quiz generator using Claude Code and Qwen.

6 min · AI-Assisted · Education · Python

How I built a comprehensive mortgage cost calculator using Claude Code and Qwen.

5 min · AI-Assisted · Finance · TypeScript

How I built a comprehensive stock tracking and analysis tool using Claude Code and Qwen.

6 min · AI-Assisted · Finance · Python