INDEPENDENT SOFTWARE CONSULTANT

Senior engineering, without the full-time hire.

I’m Esteban Selaya, a mobile engineer who went full-stack. DevStrike is my one-person practice: Android and web product engineering, production infrastructure, and AI-assisted delivery. You talk to me, and I write the code.

MOBILE · FULL-STACK · FINTECH · SECURITY · AI-ENABLEMENT

01PROBLEM → OUTCOME

You need senior engineering judgment, not another resource on a spreadsheet.

Most early-stage teams hit the same wall: the roadmap needs a Plaid integration, a security review, or a real CI/CD pipeline, but hiring a full-time senior engineer for a 6-week problem doesn’t pencil out. A generic agency isn’t much better. You end up re-explaining your stack to whoever’s on the bench that week.

That’s the gap I fill. You get me, directly, for exactly the scope that needs senior attention — an assessment, a sprint, or an ongoing fractional slot. I’m the one who opens the pull requests.

WHAT THAT LOOKS LIKE IN PRACTICE

  • A non-technical founder with a real foundation — auth, database, deploys — and an AI workflow configured to build on it safely, instead of a prototype nobody can maintain.
  • A Plaid integration that survives a signature-verification review and doesn't double-process a retried webhook.
  • A security posture you can describe honestly in a due-diligence call, because the encryption, access control, and audit trail actually exist in the code.
  • An engineering team that ships faster with AI coding agents without the bug count rising to match.
03PROCESS

How an engagement runs

  1. 01

    Send project details.

    A short form: what you're building, what's broken, and what “done” looks like. You don't have to sit through a sales call to get a reply.

  2. 02

    Scoping call (30 min).

    Thirty minutes to confirm the actual problem behind the symptom and pick the engagement shape: paid assessment, fixed-scope sprint, or fractional retainer.

  3. 03

    Fixed-scope work, visible as it happens.

    Pull requests, not status decks. You watch commits land against the agreed scope.

  4. 04

    Handoff or continue.

    Every engagement ends with documentation a teammate can pick up without me in the room. Retainers renew because you want them to, never because leaving is painful.

04CREDIBILITY

The systems I run myself

Everything below is from my own production system. These are the patterns I bring to client work, because I live with them:

  • A production fintech pipeline: Plaid transactions, investments, and liabilities flowing into an encrypted database, with signature-verified webhooks and idempotent processing. It survives retries, replays, and reviews.
  • Security I'd hand an auditor: column-level encryption with key rotation, passkey and MFA sign-in, an append-only audit log on every sensitive action, and a deploy pipeline with vulnerability scanning and a tested recovery runbook.
  • AI coding agents, daily, in production: Claude Code with custom hooks, skills, and guardrails, running safely against a codebase that handles real financial data. The full detail lives on the service pages.

You won’t find client logos here; most of my engagements are under NDA. What you get instead is my own system, described accurately, so you can judge the engineering for yourself rather than take a testimonial’s word for it.

05FAQ

Frequently asked questions

Q1What kind of engagements do you take on?
Three shapes: a paid assessment (1–2 weeks, fixed fee, ends with a written scope and prioritized findings), a fixed-scope sprint (2–6 weeks against an agreed deliverable), or a fractional retainer (a set number of days per month, ongoing). Most new relationships start with an assessment or a small sprint so both sides can confirm fit before committing to a retainer.
Q2Are you fractional, or do you do fixed-scope projects?
Both. The engagement type follows the problem: a well-defined integration or hardening pass fits a fixed-scope sprint, and ongoing product development fits a fractional retainer. We agree on the shape during the scoping call, before any work starts.
Q3Do you work with early-stage startups, or only funded companies?
Both, with one caveat: I'm a solo practice, so the fit is best where the problem needs senior judgment on a bounded scope rather than a team staffed up under one PO. Pre-seed teams shipping their first Plaid integration and Series B teams doing a pre-diligence security pass are both common.
Q4Do you write code yourself, or just advise?
Code. Every engagement produces pull requests, migrations, or infrastructure changes, not a slide deck telling someone else what to build. Advisory-only work (architecture review, security assessment) is available, but it's the exception, not the default.
Q5How do you handle security-sensitive data during an engagement?
The same way I handle it in my own systems: least-privilege access to what the engagement actually needs, no plaintext secrets in chat or email, and a clean handoff of any credentials at the end. If your project involves financial or personal data, we cover that explicitly during scoping. The Security & Infrastructure Hardening page describes the patterns I apply.
Q6What does "book an intro call" actually commit me to?
Nothing. It's a 30-minute conversation to understand the problem and tell you plainly whether I'm a good fit, including saying no if I'm not. I don't send a proposal until that conversation (or an equivalent written exchange) has happened.

GET IN TOUCH

Have a project that needs senior engineering judgment?

Tell me what you're building and what's blocking it. I'll reply within one business day, either with times for a scoping call or a straight answer that it's not a fit.