Long-form writing about AI infrastructure, implementation strategy, and how to make automation useful in day-to-day operations.
builtwithjon.com now exposes a free, public MCP server, so Claude and other AI clients can run the Hidden Profit scorecard, calculate workflow leaks, and search 96 worked use cases without leaving the conversation.
Claude connects to one Gmail or Outlook account at a time. Here are the practical ways to read, triage, share, and send across more than one inbox.
General contractors do the extra work, then lose the paper trail that would get them paid for it. The math on undocumented change orders for a $400K job, and the fix that starts before the invoice, not after.
Patient no-show rates hit roughly 6.81% in 2023 and are still climbing at most practices. The math on what an empty chair costs a 3-provider clinic, built from your own visit value instead of a borrowed per-appointment figure, and the reminder-and-waitlist fix that closes the gap without new staff.
Buyers work with whoever answers first, not whoever's the best agent in the market. The math on what a slow reply costs a real estate agent or small brokerage, and the reviewed follow-up loop that fixes it without a call center.
Every active real estate transaction is a moving stack of disclosures, addenda, and deadlines spread across a dozen parties. The transaction-coordination week has no clean hours-per-week number worth quoting — but the math on your own file load shows exactly where it's crowding out client-facing work.
On the average construction job, roughly a third of the work week goes to hunting for information, resolving conflicts, and redoing work that miscommunication caused — not to building anything. The math on what that costs a general contracting crew, and the fix that starts with where information lives.
60% of law firms don't respond to email inquiries at all, and phone answer rates have fallen since 2019, per a 2024 secret-shop study of 500 firms. The math on what a slow first response costs a professional services firm, and the fix that doesn't require a receptionist.
56% of small businesses are owed money on unpaid invoices, averaging about $17,500, and nearly half have invoices 30+ days overdue. For a home services shop, that gap doesn't stay an accounting problem — it becomes a Friday payroll problem. The math on a 6-truck shop, and the fix that doesn't require a bookkeeper.
A DM or inquiry from your content is a lead the same as any other, and it decays the same way — the odds of qualifying it drop roughly 21× once response time slips from 5 minutes to 30. How a one-person expertise brand keeps the reply fast without breaking the recording block.
A leasing inquiry that sits for even a day doesn't wait — the prospect applies wherever answered first, and the vacancy clock keeps running on the unit that lost them. The math on what slow response actually costs a property manager, and the fix that starts with triage, not a bigger leasing team.
When a homeowner's call hits voicemail, most don't leave a message — they call the next contractor on the list. The math on missed calls for a plumbing, HVAC, or electrical shop, and the first fix that doesn't require hiring a dispatcher.
A new patient calls, gets voicemail, and calls the next practice on the list — research on response speed shows the odds of ever qualifying that lead drop roughly 21-fold once response time slips from five minutes to thirty. The compliance-aware intake loop that answers fast without letting a bot near patient information, and the lifetime-value math on why it matters.
For law, accounting, and consulting firms, the expensive leak is expert hours going to non-billable admin: reconstructing documents, chasing status, drafting engagement letters from scratch. How to audit it in one week, and the math on fixing the workflow before adding headcount.
For a solo coach or creator, the hours that disappear aren't spent making the thing — they're spent repackaging it into six formats, chasing the right crop size, and re-scheduling posts by hand. The math on what that packaging tax costs your own week, and where to actually cut it.
Members and clients rarely announce they're leaving a one-person brand — they just stop renewing, stop opening emails, and let the subscription lapse. Reliable churn benchmarks for coaches and creators don't exist, so the real fix is a renewal and offboarding workflow that catches the silence before it becomes a pattern, not a number to chase.
Quotes get squeezed into the gaps between jobs, sent whenever there's a spare ten minutes, and followed up on rarely. There's no reliable trade-specific hours figure for this — what's proven is that response speed itself decides who gets the job. The math on what a delay costs, using your own numbers.
Late rent isn't one bad month — it's a chase that restarts from zero every month, on a portion of every portfolio, whether or not anyone scheduled the time for it. What the chase actually costs a property manager, and the fix that doesn't require a collections hire.
Billable utilization at professional services firms fell to 68.9% in 2024, and 56% of small businesses are owed an average $17,500 in late payments, per 2024–2025 industry benchmarks. The fee clock starts when the engagement letter goes out, not when you shook hands — here's the math on what a slow letter costs.
Small property management firms don't run out of hours to one big fire — they run out of hours to the same five tenant situations repeating every week. What the loop actually looks like, and the workflow fix that doesn't require a new coordinator hire.
Contractors lose bids in the gap between sending the estimate and following up, not in the pricing. The math on what a 24-hour follow-up delay costs a general contractor, and the one fix worth making first.
The AMA's 2024 Prior Authorization Physician Survey found physicians and staff spend roughly 13 hours a week on prior auth — nearly two clinical days. The math on what that costs a 3-provider clinic, and the workflow fix that keeps every clinical judgment with a human, inside HIPAA-compliant tools with a signed BAA.
Commission is paid at closing, not invoiced along the way, which hides the real leak: the follow-through gap between contract and close, where an unchecked contingency or a missed update quietly costs a deal. No fall-through rate to quote — just where to look in your own pipeline.
The full archaeology of building one shared memory for three AI agents: 21+ systems scored on paper, 7 actually deployed and torn down, a 180-probe eval harness, and the failures I almost shipped (a 27% baseline that was lying, a V8 ceiling that killed semantic search, and a fancy engine that lost 0-to-6 to a SQLite file).
A folder of markdown is not AI memory. Six steps turn an Obsidian vault into something an agent can query, including the MCP layer everyone skips.
A practical guide to AI-assisted job search for executives and operators. Tracker + LLM + manual review — the workflow that works, plus 11 open-source repos, X threads to inspect, and a copy-paste prompt pack.
The Center of Excellence arc from RPA is repeating in AI agents: point solutions, organic spread, centralized governance, then federated execution. If you lived through RPA, you already have the map.
Autonomous revision loops regressed every technical draft I tested. The obvious suspect was self-review: each model critiquing its own work. So I swapped in Claude Sonnet 4.6 as a strong external reviewer and reran the matrix. Every model still regressed. The bug isn't the reviewer. It's the loop.
Same seven models. Same three-judge panel. Same day. Change the writing register from technical build-log to reflective essay and the leaderboard inverts. 'Best model' is the wrong unit.
Two open-weight cloud models shipped generation upgrades this month. Both are now within 5 quality points of GPT-5.4 on writing tasks, half the cost, and faster on tool-use workloads. The frontier is getting crowded.
Qwen3.6-35B-A3B running locally on my M4 hit 40 tokens/sec on long writing — faster than GPT-5.4, Kimi K2.6, and MiniMax M2.7. It also scored 20+ quality points lower than all three. The local-cloud speed story is solved. The quality story is not.
I collected 99 YouTube transcripts, 3 audiobook transcripts, 15 guest podcast transcripts, and built a voice analysis to create a Hormozi advisor I can chat with through Claude Projects. Here's the full process.
Jules v4.2 adds social engagement and config hygiene skills while retiring RTK and trimming hooks from 5 to 3. More capabilities, fewer moving parts.
A thin hierarchy beats a giant CLAUDE.md. Where system knowledge actually belongs.
I built 24/7 AI infrastructure on Claude Code before Anthropic shipped it natively. Building ahead of the platform was the easy part. Recognizing when to delete what still worked was the real skill.
Claude Code's autoMemoryDirectory is simple to set, but each environment (CLI, self-hosted Docker, Claude Code Web) needs its own configuration path. Here's how to wire it up everywhere, with the sync strategy that ties them together.
Migration assessment: 7 of 11 cron jobs eliminated, 3 move cleanly, 2 restructure. The single blocker is persistent messaging. Decision framework for evaluating your own self-hosted AI automation.
How a Claude Code skill with challenging questions, decision science frameworks, and personal context integration turned a coding tool into a genuine thinking partner for life and business decisions.
A deep dive into running Claude Code as a background service. Three-phase orchestrator, parallel dispatch, memory synthesis safety rails, and every claude -p gotcha I discovered the hard way.
How a 300-word personality spec in CLAUDE.md transformed Claude Code from a tool into a collaborator. The implementation, the evolution, and why directness is the highest-leverage override.
The 5-layer value stack model for solo founders using AI. Infrastructure, workflow, tasks, agent management, strategy. The real leverage isn't doing more. It's doing different work.
Flow-state builders don't need less AI. They need an off-ramp built into the tool itself. Here's how I used Claude Code to solve the problem Claude Code creates.
Real usage data from running Claude Code across two machines on one Claude Max account. Five days of billing block analysis with ccusage, promo window math, and practical takeaways for managing usage.
Jules v2 container infrastructure: 8-phase boot sequence, 7 scheduled jobs, Slack daemon, credential flow, and the architectural patterns that make an AI agent truly autonomous.
NemoClaw dropped yesterday. My first reaction: I already have most of this. A practitioner's comparison of Docker sandbox isolation, CLAUDE.md policy controls, and the one real gap (inference routing) that NemoClaw actually adds.
Claude Code's Explore agent rediscovers your workspace from scratch every time. A pre-computed structural index cuts that to one read. Full code for both the index generator and the custom agent.
I ran my 116-configuration Claude Code setup against @darnoux's 10-level mastery framework. Level by level breakdown with practical details at each stage.
A structured Claude Code personality with voice registers, decision authority, standing orders, proactive behaviors, and a builder's trap detector. Full implementation shared.
A full audit of a production Claude Code workspace: 6 CLAUDE.md files, 29 skills, 5 agents, 22 rules, 8 hooks, 43 Makefile targets, and more. Every configuration explained.
A file-based async messaging system for Claude Code. Drop a request in a markdown file from your phone, get results back in 20 seconds. No terminal needed.
The wrap-up skill commits code, updates memory, runs a self-improvement review, and flags publishable content. Four phases that make every session compound.