Available for select projects
I build what your business actually needs.
Internal tools, AI integrations, and the production systems small teams can't afford to get wrong — scoped, built, and handed off by one person.
Categories
- 01
Brief
Read the intake. Decide fit.
- 02
Scope
Reduce to what ships.
- 03
Build
Smallest strong version.
- 04
Hand off
Docs, deploy, env. Clean exit.
The intake is structured because the work is structured. Scope control is part of the deliverable.
Who this is for
Fit is shaped by the problem, not the buzzwords.
Right fit
- A spreadsheet, email, and Zapier chain that breaks every time the team grows.
- Your first internal tool — beyond what no-code can hold, beneath what an agency will quote.
- An AI integration that needs real auth, real data, and real cost control. Not a chat box.
- A CRM, billing, and support stack that doesn't talk to itself — and the founder has been quietly dreading it.
- A small business-critical website where the wrong framework or host costs more to undo than to build right.
Not a fit
- Cheapest-possible execution, no judgment expected.
- Pre-revenue idea-stage scoping disguised as a project.
- Agencies looking for white-label dev capacity.
- "Make me a dashboard" tickets without business context.
- Procurement-heavy enterprise process with MSA / SOC2 questionnaires.
What I build
A few connected categories. Not everything.
Each row carries a definition, a fit signal, and a real example. You can self-identify before sending the brief. I can triage when it arrives.
Internal tools & admin systems
Lead intake, ops dashboards, internal CRMs, customer portals — the systems your team uses every day.
For you if: You need a real internal tool, not another spreadsheet, and the team is too small for an in-house engineer.
Not for you if: A no-code stack already does the job and the data fits inside it.
e.g. Multi-step intake → Postgres → auth-gated admin with status workflow.
AI integrations that work in production
Models wired into real data, real auth, and real cost control. Built to ship, not to demo.
For you if: You have a workflow where an LLM, retrieval, or a custom pipeline could actually replace human steps.
Not for you if: You want a chat box bolted onto a marketing site to "have AI on the site."
e.g. RAG pipeline over support history, gated by tenant, with token-cost guardrails and a fallback path.
System integrations & data plumbing
The middle layer that makes your business actually run — CRM ↔ billing ↔ support ↔ product.
For you if: You can name three systems that should be talking to each other and aren't.
Not for you if: Off-the-shelf Zapier/n8n flows already cover it and stay reliable under load.
e.g. Stripe events → CRM updates → support context → internal alerting, with replay and idempotency.
Business-critical small websites
When the wrong framework, host, or CMS will cost more to undo than to build right the first time.
For you if: It's a real business surface — leads, payments, regulated content — and downtime or migration cost is real money.
Not for you if: You need a brochure site and the cheapest agency in your network can ship it next week.
e.g. Next.js on Vercel with structured intake, GA4, GSC, sitemap, canonical, and a clean handoff path.
Technical second opinion & scoping
Short engagement, structured deliverable. A real read on the problem before money gets spent.
For you if: You're about to commit to a build and want senior judgment on scope, stack, and trade-offs first.
Not for you if: You want a free strategy call disguised as scoping.
e.g. Two-week paid scoping — written assessment, recommended path, explicit cuts, and a build/no-build call.
How I work
Disciplined intake. Honest scope. One person shipping.
The intake is structured because the work is structured. Scope control and restraint are part of the deliverable, not a pre-sale ritual.
- 01
Brief
You send a structured project brief through the intake form. I read every brief personally. I reply with a real read on fit — yes, no, or "not yet, here's why."
- 02
Scope
We reduce the problem to what actually needs to ship. I tell you what to cut. The smallest strong version usually beats the bigger plan you came in with.
- 03
Build
I ship the smallest strong version. Production-grade from day one — auth, env vars, observability, cost controls, and a deploy path. Not a demo.
- 04
Hand off
Documentation, deploy notes, env, on-call expectations. Clean exit. You end up with a system you can run without me — or a relationship where I keep maintaining it on clear terms.
Why one person
The absence of an agency team is intentional.
One person, end-to-end.
The person you brief is the person who ships. No handoff between strategy, build, and operate. No account manager. No "let me check with the team."
I decide what not to build.
The default move is reduction. A 200-line solution that fits the actual problem usually beats the 2,000-line platform that doesn't.
Production discipline, not demos.
Auth, env vars, observability, cost — load-bearing from day one. The systems I ship are run by real teams under real conditions, not pitched on a stage.
Direct contract.
You hire me, you work with me. No project coordinators, no fractional CTO arrangement, no "we'll build a team for you" upsell. One person, intentionally.
Restraint as a feature.
I'll tell you when the right move is buying the SaaS instead of building it. The work I take on is the work that actually deserves a custom build.
Proof
This site is one of mine.
Most consultants describe their work. I'd rather show you. The site you're reading runs on the same discipline I bring to client projects — every surface real, every integration in production.
- Structured 4-step intake → Neon Postgres
- Server actions for status and notes, no client-side writes
- Auth-gated admin dashboard with status workflow
- Defense-in-depth on admin surfaces
- GA4 + Google Search Console verification
- Sitemap, robots, canonical host normalization
- Vercel production deploy, env-gated routes
First client case studies will publish as engagements ship and permissions/artifacts come available. I'd rather show real work than fake proof.
Questions worth answering
The objections most prospects don't voice.
- Why work with one person instead of an agency?
- Because the person you scope it with is the person who builds it — and the person who builds it is the person who maintains it. No handoff, no telephone-game between strategy and engineering. The trade-off is throughput. I take on fewer projects than an agency could. That's the point.
- What if my project isn't a fit?
- I'll tell you. Most projects aren't a fit — for me, for someone else, or for now. If yours isn't, you'll get an honest read on why and, where I can, a pointer to who or what is. The intake is short for that reason: if it can't be answered in 8 minutes, the project usually isn't ready to start.
- Do you work with non-technical founders?
- Yes — most clients are founders or operators, not engineers. The intake is structured so you don't need technical vocabulary to describe the problem. I translate the business shape into a technical plan, not the other way around.
- Can you take over something half-built?
- Often, yes. I'll need to read what's there before I can commit. Some projects are easier to finish than to inherit; some are the opposite. If I don't think a takeover is the right call, I'll tell you that too.
- What does a typical project cost?
- Pricing comes back with the brief response, not before. The brief gives me enough context to give you a real range — not a generic one. Engagements vary widely depending on scope, timeline, and what actually needs to be built.
- What happens after I submit the brief?
- I read it personally. You get a real reply — fit assessment, scope read, and either a path forward or an honest no. No calendar gate, no sales call before the problem is clear. If we move forward, the next step is a written scope, not a kickoff meeting.
Next step
Submit a brief.
8 minutes. 4 steps. Enough context for a real answer — not a sales call. I read every brief personally.
Submit a project briefNo calendar gate. No sales pitch. If your project isn't a fit, I'll tell you why.