Work
ProjectsCase studies, shipped work, and prototypes.Open SourceGitHub, npm, PyPI, and public packages.Hackathons & CompetitionsHackathons, contests, and rapid-build work.Depth
JourneyCareer timeline, roles, and education.AchievementsScholarships, rankings, and milestones.Courses & CertificationsStructured learning, courses, and credentials.SkillsTechnical capability map.This website · 02
Browse this website posts across architecture, delivery, product systems, and the public writing footprint connected to this site.
Zero-downtime deploys on a cheap VPS
Field notes
485 published articles. Essays written here open in place; older tutorials take you back to the site where they first appeared.
A rate-limit-by-rate-limit, price-tier-by-price-tier comparison of what HubSpot, Salesforce, and GoHighLevel actually let an autonomous agent do.
ArticleQuote-to-cash breaks down in the handoffs, not the steps. Here's a stage-by-stage look at where AI agents actually close the gaps between CPQ, contracts, and cash.
ArticleMost RevOps stacks are not too small, they are too tangled. A category-by-category audit for finding overlap, dead tools, and integration debt before you buy anything new.
ArticleA rate-limit-by-rate-limit, price-tier-by-price-tier comparison of what HubSpot, Salesforce, and GoHighLevel actually let an autonomous agent do.
TutorialA production-grade walkthrough of connecting an AI agent to HubSpot: private apps, real scopes, actual rate limits, webhooks, and signature verification.
CareerWhen researchers put monitoring software on 20 professional developers' machines for 220 work days, coding came out at 21% of the day. Not because those developers were slacking — because the other 79% is the job. AI automates a slice of the 21%.
TutorialREST, Bulk API 2.0, Platform Events, and Composite API each solve a different agent-integration problem. Here's which to use, with the real governor limits.
TutorialHow GoHighLevel's sub-account architecture, API v2, and snapshot system actually work — and where the workflow builder ends and code needs to start.
ArticleAutomations fail downstream of the data model, not the workflow logic. Here are the object, picklist, and relationship mistakes that break every fix built on top.
ArticleData hygiene policies rely on reps remembering rules under quota pressure. AI agents enforce the same rules at the point of entry, automatically.
AI / MLWhen an LLM stops answering questions and starts calling APIs, the attack surface moves from bad text to real-world actions. Here's what actually changes.
TutorialA wildcard certificate can't be proven the way a normal one is — there's no single page to fetch for a domain that doesn't exist yet. That's why it needs a DNS record instead of an HTTP request, and why Cloudflare's API has to be involved at all.
ArticleOff-the-shelf CRMs gate agent-grade API access behind pricing tiers, not feature lists. A concrete framework for knowing when a custom build actually pays for itself.
TutorialPipedrive and Close both automate well below Salesforce pricing. Here's what their APIs, webhooks, and native automation actually let a small team build.
Cloud / DevOpsVercel gives you monitoring whether you ask for it or not. A VPS gives you a blank terminal and the assumption you'll figure it out. Here's what to actually watch on a self-hosted box, with tools that cost nothing, and why watching from the box itself is the one setup that will lie to you.
AI / MLRecord linkage, fuzzy matching, and blocking are decades-old techniques. Here's how an AI agent wires them into a CRM that stays clean without a quarterly cleanup project.
TutorialData migrates. Automation logic doesn't. Here's the field mapping, dedup, and workflow-rebuild sequence that keeps routing, scoring, and sequences intact through cutover.
CareerThe useful question isn't what AI can do — it's what it structurally cannot. Veracode ran 100+ models across 80 tasks and 45% of the output carried an OWASP Top 10 vulnerability, with larger models no better than small ones. That failure has a shape, and the shape tells you what to learn.
ArticleSmall teams can outpace enterprises by treating AI agents as workforce multipliers. Here's how to structure roles, ownership, and accountability.
ArticleCut through hiring myths. Compare fully-loaded FTE costs against contractors and AI agents with real math—not benchmarks, but the numbers that matter for RevOps teams.
TutorialTreat AI agents as employees: document their scope, provision minimal-privilege access, and implement review gates. This framework ensures agents operate safely while retaining autonomy.
CareerA RevOps automation engineer builds scalable systems connecting sales tools, fixes revenue pipeline leaks, and automates GTM workflows. They blend engineering rigor with sales operations expertise.
ArticleStop guessing. Real AI agent performance hinges on measurable metrics—success rate, latency, token efficiency, and cost per task. Here's how to review what actually matters.
CareerNot every role is ready for agent replacement today. Learn which functions automate first—and why repeatable, data-driven tasks matter more than job titles.
AI / MLRunning multiple AI agents in production requires observability, alerting, and graceful human handoff patterns. Here's the architecture that separates chaos from control.
CareerRichard Cook's central claim is that catastrophe requires multiple failures — no single fault is ever enough. That's a statement about relationships between components, which is exactly the information a code generator never receives. Syntax got commoditized. The relationships did not.
CareerEntry-level hiring at the tech majors is down 65% since 2019 and Stanford measures a 19% employment gap for 22-to-25-year-olds. But an LSE paper covering 243 million hires found that when you control for remote work, the AI effect largely vanishes. The cause matters, because the two have opposite fixes.
TutorialA complete guide to implementing an outbound voice calling system for sales using modern voice AI platforms. Real implementation patterns, architecture decisions, and platform API walkthroughs.
CareerJudgment is the ability to be right about things that cannot be checked yet. Cheap generation raises the volume of decisions and lowers the cost of each one, which sounds like it devalues judgment. It does the opposite, and DORA's stability data is the receipt.
ArticleVoice AI promises 24/7 support and cost savings. Real deployments show 40-60% resolution rates and a hidden liability: when customers get angry at machines, they don't just hang up.
CareerStack Overflow's 2025 survey found the top developer frustration wasn't AI being wrong. It was AI being almost right — output that compiles, looks correct, and costs you an afternoon. That failure mode moves work out of writing and into diagnosis, and diagnosis was already the expensive half.
A working notebook
“Some ideas belong here because they connect directly to projects and services.
Older tutorials still live on Medium, where people first found them. This page keeps the trail intact without pretending everything was published in one place.
How I choose a topic
I would rather publish one useful explanation from lived work than five summaries of things everyone already knows.
Step 01
A topic usually starts as the second or third time the same engineering, product, hiring, or research question lands on my desk.
Step 02
I look for a shipped system, a measured result, a failed approach, or a public artifact that keeps the argument honest.
Step 03
Most technical choices touch something else. The useful links are the ones that help a reader follow that chain without opening fifteen tabs.
Step 04
Some pieces belong on this site; others already have a life elsewhere. I keep the original destination and publishing history intact.
Topics covered
The recurring technical and strategic themes represented across this archive.
A quick orientation
The practical details behind this mixed archive.
Those pieces were published there first. I would rather send you to the original article—with its date, responses, and history—than create a duplicate here.
Thirty-five posts are enough to become noisy. The filters separate original essays from outside tutorials, and permanent page links make it easier to pick up where you left off.
Usually. The essays here tend to connect a technical decision to the project, service, or research context behind it. External posts are often narrower, practical guides.
Please do. Real questions are where the best pieces start. Send the context, what you have already tried, and the part that still feels unclear.
Next step
You do not need a polished brief. A bottleneck, an uncertain architecture decision, or a product idea that refuses to become concrete is enough to start.