Cost, time zone,
and how I work.

It depends on scope, but I price in three models so the number is predictable before any work starts. Fixed-scope projects — a defined build with a clear outcome — are quoted as a single price after a short discovery call, so you know the cost up front and carry no estimation risk. Monthly retainers suit ongoing product work, where you book a set amount of senior capacity each month. For focused work, audits, or advisory there's a day rate, from US$1,000/day.

What actually moves the price is real complexity: how many integrations, whether there's AI, LLM, or data work involved, security and compliance requirements, and how much already exists versus building from zero. I'd rather scope it honestly with you than quote a meaningless number sight-unseen — the discovery call is free, and you leave it with a written estimate.

Fixed-price for defined builds, day rate or retainer for open-ended work — but not hourly. I don't run a meter. For well-defined projects, a fixed price means you carry no risk on the estimate and I'm rewarded for being efficient rather than slow. For evolving work, a day rate or monthly retainer keeps things flexible without an open-ended bill. I'll recommend whichever model genuinely fits your project on the first call — not the one that's best for me.

A short, free intro call — you tell me what you're building or what's broken, and I tell you honestly whether I'm the right fit and how I'd approach it. From there, many engagements begin with a two-week paid diagnostic: I read the code, interview the team, and send back a memo with a concrete plan, risks, and a staffing recommendation. About 70% of diagnostics roll into a 3–12 month engagement; the rest take the memo and run it themselves — a clear memo is a valid deliverable either way.

NDA-first is fine — I'll sign yours or send mine, whichever is faster. And if your site or app is actively on fire, the urgent triage track is faster: diagnosis is free, and you only pay if you ask me to do the fix.

AI-native products where the model's experience is a first-class concern; brownfield migrations where the team is stuck; editorial and civic platforms that have to last a decade. Broadly: anything where the engineering actually matters and you'd rather have one senior generalist than a team of four mid-levels.

Not a fit: staff augmentation, pure prototype work, or being the only senior voice in a large, dysfunctional org.

Solo most of the time — augmented by a tight multi-agent workflow: planner, engineer, reviewer, QA, and devops sub-agents via Claude Code, with hard process gates between stages. For larger engagements I bring in a small bench of long-term collaborators — a senior designer, a backend partner, a mobile specialist — assembled per project rather than kept on retainer.

I'm based in Lima, Peru, and work remote-first with clients worldwide. Lima stays on UTC-5 all year — Peru doesn't observe daylight saving — so I overlap a full US working day: aligned with US Eastern through the winter and US Central across the summer. In practice that means real-time collaboration with North American teams, not the overnight handoffs you get from a 10+ hour offset. I work in fluent English and Spanish.

Yes — most of my work is remote for clients outside Peru, including in the US and Europe. As a nearshore engineer in a US-overlapping time zone, I join your standups, pair when it helps, and respond within your working hours rather than the next day. I've delivered for organizations including MIT, Harvard, AWS, and Condé Nast, all remote-first.

Yes. A good share of my work is white-label senior capacity for agencies and studios — I plug into your team and your process, ship under your brand, and stay invisible to your client if that's what you need. It's a clean way to add a senior engineer for overflow, a specialist AI or Laravel build, or to cover a gap, without a full-time hire. I'm comfortable working inside your repos, your tooling, and your client communications.