Notes from running Claude Code at scale. AI-assisted development, parallel agent orchestration, and what actually ships.https://tech.bdigitalmedia.io/en-ushttps://tech.bdigitalmedia.io/blog/the-cost-of-accuracy/https://tech.bdigitalmedia.io/blog/the-cost-of-accuracy/You don't buy answers from a frontier model, you buy passes. Most real tasks take more than one, and the verification passes are where the budget actually goes.Tue, 26 May 2026 13:35:06 GMTdevaiclaudecostagentsverificationevalshttps://tech.bdigitalmedia.io/blog/friction-is-the-feature/https://tech.bdigitalmedia.io/blog/friction-is-the-feature/The four-practice discipline I built after eighteen months of heavy AI use. Without deliberate friction, AI compounds capability and dependence.Tue, 12 May 2026 13:34:46 GMTaiclaude-codedisciplineskill-atrophyoverreliancehttps://tech.bdigitalmedia.io/blog/harness-is-the-bottleneck-five-prompts-broke-on-opus-4-7/https://tech.bdigitalmedia.io/blog/harness-is-the-bottleneck-five-prompts-broke-on-opus-4-7/Adaptive thinking and a literalness shift turned the Opus 4.7 prompt into a multi-dimensional contract. Half your prompt library signs old terms.Sun, 10 May 2026 19:31:38 GMTaiclaudeprompt-engineeringevalshttps://tech.bdigitalmedia.io/blog/llm-fundamentals-agentic-loop/https://tech.bdigitalmedia.io/blog/llm-fundamentals-agentic-loop/The agentic loop is a while loop that checks stop_reason, executes tools, and returns results. It is 10 lines of code that depend on everything from posts 0-8.Sun, 10 May 2026 14:36:54 GMTaillm-fundamentalshttps://tech.bdigitalmedia.io/blog/llm-fundamentals-from-loop-to-agent/https://tech.bdigitalmedia.io/blog/llm-fundamentals-from-loop-to-agent/Scaling an agentic loop to production means managing cost, context, and complexity. A roadmap: prompt caching, model routing, compaction, subagents, MCP.Sun, 10 May 2026 14:36:54 GMTaillm-fundamentalshttps://tech.bdigitalmedia.io/blog/llm-fundamentals-tool-use/https://tech.bdigitalmedia.io/blog/llm-fundamentals-tool-use/Tool use lets Claude call functions you define. You ship schemas, the model picks when to call, your code runs the operations. Prerequisite for agents.Sun, 10 May 2026 14:36:54 GMTaillm-fundamentalshttps://tech.bdigitalmedia.io/blog/llm-fundamentals-structured-output/https://tech.bdigitalmedia.io/blog/llm-fundamentals-structured-output/Structured outputs guarantee JSON schema conformance at the API level, which removes an entire category of parsing and validation code from your pipeline.Fri, 08 May 2026 11:33:25 GMTaillm-fundamentalshttps://tech.bdigitalmedia.io/blog/karaoke-audio-by-voice-200-miles-out/https://tech.bdigitalmedia.io/blog/karaoke-audio-by-voice-200-miles-out/Built a word-by-word audio sync feature for a blog post without touching my laptop. Voice on the phone, agents on the laptop, every change shipped to prod.Sun, 03 May 2026 15:21:00 GMTaiclaude-coderemoteagentshttps://tech.bdigitalmedia.io/blog/llm-fundamentals-extended-thinking/https://tech.bdigitalmedia.io/blog/llm-fundamentals-extended-thinking/Extended thinking gives Claude a reasoning scratchpad before responding. It costs output tokens but measurably improves accuracy on hard problems.Sat, 02 May 2026 15:01:20 GMTaillm-fundamentalshttps://tech.bdigitalmedia.io/blog/working-isnt-repeatable/https://tech.bdigitalmedia.io/blog/working-isnt-repeatable/Frontier LLMs are not deterministic, even at temperature=0. Same prompt, five different answers. Why it compounds when you stack LLMs, and how to recover.Wed, 29 Apr 2026 13:24:25 GMTdevaiclaudeevalsstochasticityhttps://tech.bdigitalmedia.io/blog/llm-fundamentals-prompt-engineering/https://tech.bdigitalmedia.io/blog/llm-fundamentals-prompt-engineering/As models get smarter, the value shifts from crafting prompts to designing the system around them. Where the prompt ends and the harness begins is the question.Mon, 27 Apr 2026 22:03:07 GMTaillm-fundamentalshttps://tech.bdigitalmedia.io/blog/exponential-ai-snowball-case-study/https://tech.bdigitalmedia.io/blog/exponential-ai-snowball-case-study/A 16-hour Claude Code session shipped a long-form post, three SVG diagrams, a CI script, and four legacy fixes from a seven-word instruction.Mon, 27 Apr 2026 22:02:36 GMTaiproductivityclaude-codecase-studyhttps://tech.bdigitalmedia.io/blog/rag-on-call-investigator-mastra-bedrock/https://tech.bdigitalmedia.io/blog/rag-on-call-investigator-mastra-bedrock/Long context did not kill RAG for incident response. Per-section indexing in pgvector, hybrid .docx extraction, an iterative Mastra agent loop on Bedrock.Mon, 27 Apr 2026 22:02:36 GMTairagmastraawsbedrockincident-responsehttps://tech.bdigitalmedia.io/blog/claude-ai-cowork-claude-code-comparison/https://tech.bdigitalmedia.io/blog/claude-ai-cowork-claude-code-comparison/Three Claude surfaces, three context windows, three different ways of knowing about your files. Confirmed only from Anthropic's own documentation.Sat, 25 Apr 2026 21:15:31 GMTaiclaudeclaude-codehttps://tech.bdigitalmedia.io/blog/llm-fundamentals-messages-api/https://tech.bdigitalmedia.io/blog/llm-fundamentals-messages-api/Every Claude API call is stateless. You send roles, messages, and the full conversation history each time. Here is how that structure works.Sat, 25 Apr 2026 17:18:01 GMTaillm-fundamentalshttps://tech.bdigitalmedia.io/blog/cc-context-monitor-statusline/https://tech.bdigitalmedia.io/blog/cc-context-monitor-statusline/A one-command Claude Code plugin renders three progress bars: context fill, 5-hour session burn, 7-day weekly quota. No ccusage required on CC v2.1.x+.Mon, 20 Apr 2026 12:00:00 GMTdevaiclaudeclaude-codepluginsshellhttps://tech.bdigitalmedia.io/blog/evaluating-a-claude-skill-before-shipping/https://tech.bdigitalmedia.io/blog/evaluating-a-claude-skill-before-shipping/Six phases to ship a Claude skill like software: behavioral eval, packaging, review, CI, install verification, governance. Each catches a different bug.Sun, 19 Apr 2026 16:00:00 GMTdevaiclaudeworkflowevalpluginshttps://tech.bdigitalmedia.io/blog/ai-code-review-fix-patterns/https://tech.bdigitalmedia.io/blog/ai-code-review-fix-patterns/Part 3 of 3. Four structural fix patterns that move an AI reviewer from ignorable to trustworthy, plus the meta problem of using AI to audit AI.Sun, 19 Apr 2026 12:00:00 GMTdevaiclaudeworkflowleadershiphttps://tech.bdigitalmedia.io/blog/designing-ai-pr-review-claude-skills/https://tech.bdigitalmedia.io/blog/designing-ai-pr-review-claude-skills/Part 1 of 3. Building a PR reviewer as structured Claude skills: anatomy of a review skill, why markdown beats prompt blobs, what makes iteration possible.Sun, 19 Apr 2026 01:26:02 GMTdevaiclaudeworkflowleadershiphttps://tech.bdigitalmedia.io/blog/diagnosing-ai-review-false-positives/https://tech.bdigitalmedia.io/blog/diagnosing-ai-review-false-positives/Part 2 of 3. The cry wolf failure mode, the trust threshold step function, and a four-step cluster analysis that surfaces root causes, not symptoms.Sat, 18 Apr 2026 12:00:00 GMTdevaiclaudeworkflowleadershiphttps://tech.bdigitalmedia.io/blog/creatine-cognitive-performance-developers/https://tech.bdigitalmedia.io/blog/creatine-cognitive-performance-developers/Your brain burns 20% of your body's energy. Creatine refills brain ATP 40x faster than any other pathway. What that means for devs at their keyboards.Wed, 15 Apr 2026 12:00:00 GMTdevaihealthworkflowhttps://tech.bdigitalmedia.io/blog/llm-fundamentals-context-windows/https://tech.bdigitalmedia.io/blog/llm-fundamentals-context-windows/Context windows are working memory, not storage. More tokens means more cost, more latency, and accuracy that degrades as context grows.Tue, 14 Apr 2026 12:00:00 GMTaillm-fundamentalshttps://tech.bdigitalmedia.io/blog/llm-fundamentals-what-are-llms/https://tech.bdigitalmedia.io/blog/llm-fundamentals-what-are-llms/Large language models are prediction machines built on transformers. What they are, how developers use them, and where agent workflows fit in.Mon, 13 Apr 2026 12:37:41 GMTaillm-fundamentalshttps://tech.bdigitalmedia.io/blog/llm-fundamentals-text-generation/https://tech.bdigitalmedia.io/blog/llm-fundamentals-text-generation/LLMs predict one token at a time from a probability distribution. Temperature, top-k, and top-p control that distribution, not creativity.Mon, 13 Apr 2026 12:00:00 GMTaillm-fundamentalshttps://tech.bdigitalmedia.io/blog/llm-fundamentals-tokens/https://tech.bdigitalmedia.io/blog/llm-fundamentals-tokens/LLMs don't process words or characters. They process tokens, subword units that determine cost, speed, and context limits for everything you build.Sun, 12 Apr 2026 12:00:00 GMTaillm-fundamentalshttps://tech.bdigitalmedia.io/blog/superpowers-workflow-claude-code/https://tech.bdigitalmedia.io/blog/superpowers-workflow-claude-code/A structured Claude Code workflow that replaced plan-and-pray with repeatable, first-pass-clean features.Thu, 02 Apr 2026 12:00:00 GMTdevaiclaudeworkflowhttps://tech.bdigitalmedia.io/blog/proposals-portal-one-evening-claude-code/https://tech.bdigitalmedia.io/blog/proposals-portal-one-evening-claude-code/Magic link auth, progress tracking, and one-command publishing. Built with Claude Code, Astro, Cloudflare KV, and Resend from my phone.Thu, 26 Mar 2026 12:00:00 GMTdevaiclaudecloudflareastrohttps://tech.bdigitalmedia.io/blog/mastra-site-monitor-cloudflare-telegram/https://tech.bdigitalmedia.io/blog/mastra-site-monitor-cloudflare-telegram/I stood up a Mastra workflow from my phone via Telegram to monitor Cloudflare Workers. It caught bots crashing my media proxy on the very first run.Sat, 21 Mar 2026 12:00:00 GMTdevaicloudflareworkflowhttps://tech.bdigitalmedia.io/blog/pushing-to-prod-from-the-mountains/https://tech.bdigitalmedia.io/blog/pushing-to-prod-from-the-mountains/AI agents and channels integration make it possible to ship production code from anywhere: a summit, a grocery line, a dog walk. Why it works.Fri, 20 Mar 2026 14:00:00 GMTdevaiclaudeworkflowmobilehttps://tech.bdigitalmedia.io/blog/automated-self-improving-skills/https://tech.bdigitalmedia.io/blog/automated-self-improving-skills/The self-improving Claude Code skills loop from Part 1 now runs itself. A SessionEnd hook and daily GitHub Action generate improvement PRs automatically.Sun, 15 Mar 2026 12:00:00 GMTdevaiclaudeworkflowhttps://tech.bdigitalmedia.io/blog/self-improving-claude-code-skills/https://tech.bdigitalmedia.io/blog/self-improving-claude-code-skills/A self-improving loop for Claude Code skills that observes failures, finds patterns, and proposes fixes. Skills that improve over time instead of degrading.Sat, 14 Mar 2026 12:00:00 GMTdevaiclaudeworkflowhttps://tech.bdigitalmedia.io/blog/llm-seo-llms-txt-ai-discoverability/https://tech.bdigitalmedia.io/blog/llm-seo-llms-txt-ai-discoverability/How I implemented LLM SEO across 3 sites with FAQ schema, entity linking, and 14 AI crawlers, plus an honest look at what actually works and what doesn't.Sat, 07 Mar 2026 12:00:00 GMTdevaiseoastrohttps://tech.bdigitalmedia.io/blog/claude-code-remote-control-mobile-workflow/https://tech.bdigitalmedia.io/blog/claude-code-remote-control-mobile-workflow/How I use Claude Code remote control with the iOS app to ship code from anywhere. No SSH tunneling, no exposed ports, just a Max subscription.Thu, 05 Mar 2026 12:00:00 GMTdevaiclaudeworkflowmobilehttps://tech.bdigitalmedia.io/blog/seo-content-machine-claude-code/https://tech.bdigitalmedia.io/blog/seo-content-machine-claude-code/How I used Claude Code agents, context files, and slash commands to automate SEO for my Astro portfolio site: keywords, linking, and content audits.Wed, 04 Mar 2026 12:00:00 GMTdevaiclaudeseoastrohttps://tech.bdigitalmedia.io/blog/building-clip-storefront-with-claude/https://tech.bdigitalmedia.io/blog/building-clip-storefront-with-claude/Pixieset can't sell individual video clips. So I built my own storefront with Claude Code, Cloudflare R2, and Stripe, from my phone, on a Sunday morning.Sun, 01 Mar 2026 12:00:00 GMTdevastroclaudecloudflarestripevideohttps://tech.bdigitalmedia.io/blog/resume-page-and-cicd-pipeline/https://tech.bdigitalmedia.io/blog/resume-page-and-cicd-pipeline/How I built a filterable tech resume with multiple export formats and set up a proper CI/CD pipeline with GitHub Actions and Dependabot.Mon, 22 Dec 2025 14:00:00 GMTdevastroci-cdgithub-actionslinkedinhttps://tech.bdigitalmedia.io/blog/building-xmas-tracker-with-claude/https://tech.bdigitalmedia.io/blog/building-xmas-tracker-with-claude/How I went from 'I need to track Christmas gifts' to a fully deployed, PIN-protected wishlist app before my morning coffee got cold.Sun, 21 Dec 2025 12:00:00 GMTdevastroclaudeparentinghttps://tech.bdigitalmedia.io/blog/dumber-models-on-purpose/https://tech.bdigitalmedia.io/blog/dumber-models-on-purpose/Why the most capable AI isn't always the right tool, and how friction in your workflow can lead to better architecture decisions.Wed, 08 Jan 2025 12:00:00 GMTdevaiclaudeworkflowlinkedin