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.
How a mid-market firm freed up 7+ hours per week from a senior analyst by automating recurring report assembly with AI agents and structured escalation.
CareerA star count says a repo got noticed. It says nothing about the judgment behind it. Here's how to actually read a maintainer's public work.
AI / MLHow a lending platform automated 60% of compliance-sensitive support questions without sacrificing accuracy or regulatory oversight.
CareerBeing correct and winning a decision are different skills, measured by different people, on different evidence. Most engineers only ever train the first one.
CareerHow a comprehensive technical review transformed an M&A deal—reducing valuation risk and setting realistic expectations for a $40-50M acquisition.
Cloud / DevOpsThe cloud bill and the platform's slowness usually trace back to the same design decision. A good cost review fixes both at once.
ArticleHow a mid-market DTC apparel brand automated budget allocation across dozens of ad sets, cutting CPA by 28% and improving ROAS consistency by moving from weekly manual reviews to real-time agent-driven adjustments.
ArticleHow a mid-market firm freed up 7+ hours per week from a senior analyst by automating recurring report assembly with AI agents and structured escalation.
Cloud / DevOpsHow a mid-scale SaaS company reduced incident response time by 65% through unified logging, metrics, and traces—plus actionable alert tuning.
Cloud / DevOps300 million events taught me more about throughput and observability than any dashboard could. Here's where the real cost hides in large event systems, and what actually fixes it.
Cloud / DevOpsInternal platforms fail when builders optimize for structure and engineers just want to ship. Here's how to close that gap and get real adoption.
AI / MLAn AI agent decides its own next step instead of waiting for your next message. Here's what actually separates one from a chatbot or an automation script, with the workflows where the difference shows up.
CareerNot by explaining the technology better. The stakeholder isn't missing information about how the system works — they're missing a translation of what happens to something they already track if you don't get what you're asking for.
AI / MLCompare 8 leading AI agent platforms for small businesses. From no-code automation to custom frameworks, find what fits your budget and technical depth.
CareerAI agent development is immature and overhyped. Here's how to evaluate candidates properly and find engineers who can actually ship reliable systems.
CareerMost case studies are a highlight reel. The ones that build trust show the decision that could have gone wrong, and didn't.
Web DevBolt a wholesale toggle onto a consumer marketplace and it breaks within six months. Here's how one platform split B2C and B2B into separate pricing, checkout, and order systems, and grew B2B to 65% of platform revenue without touching what already worked.
ArticleAI agents excel at repetitive SDR tasks but fail at the relationship building that closes deals. Here's what actually works.
CareerGartner's 2025 buyer survey found 61% of B2B buyers would rather never talk to a sales rep at all. That's not bad news for an engineer who has to sell their own work — it's the entire opening. Credibility does what a pitch can't.
TutorialA step-by-step guide to automating CRM data entry with AI—from extracting data with LLMs to webhook-driven validation and human handoff.
CareerAsking a friend "would you use this?" measures how much they like you, not whether they'd pay. Steve Blank, Eric Ries and Rob Fitzpatrick each built a method for the same problem: telling a real signal from a polite one before you've spent the months.
AI / MLAgentic AI can deliver real ROI for small businesses, but only in specific scenarios. Here's the honest math and which use cases actually work.
CareerA hackathon prototype is good at collapsing uncertainty fast. Whether it becomes anything lasting depends on what the team does with it after the demo ends.
Cloud / DevOpsModernization fails the moment it becomes a pause button on shipping. Here's how to fix monorepos, workflows, and service boundaries without freezing delivery to do it.
CareerStripe's own data shows the gap between a top-decile solo founder and the median one has gone from 34x to 61x in four years. That gap is the whole story: a one-person software business isn't a smaller startup, it's a different asset class, and revenue per hour is the metric that proves it.
ArticlePicking a CRM is one of the biggest infrastructure decisions you'll make. Here's how to choose one that won't strangle your startup as you scale.
CareerCursor was reportedly paying Anthropic more than it collected from customers, at negative gross margin, while charging some of the lowest prices in software history. That isn't a Cursor problem. It's what happens to price and margin whenever the marginal cost of producing the good collapses — a pattern economics named forty years before AI wrote a line of code.
AI / MLA proof-of-concept takes two to four weeks. Production takes three to six months. The gap comes from integration, evaluation, and guardrails—not the model.
CareerGTM engineering is engineering for revenue teams—building integrations, automations, and internal tools that let go-to-market teams close deals faster and scale predictably.
CareerRob Walling ran the numbers on this back in 2008: 348 hours to build what you could buy at roughly 20% of the cost. The pull to build it yourself is strongest exactly where it's most expensive — payments, auth, email deliverability — because those are the parts that feel like real engineering.
CareerCB Insights reviewed public post-mortems from 431 VC-backed companies that shut down since 2023 and found 70% ran out of capital, 43% had poor product-market fit, and 19% had unsustainable unit economics. An earlier CB Insights pass, of 101 startup post-mortems, put "no market need" at 42%. Building was rarely on the list. It's the part engineers are trained for, which is exactly why it's not where startups die.
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.