Services
AI engineer
AI-assisted engineering for teams that ship — rules in the repo, small diffs, human review. Applied workflow and product integration, not research-lab ML.
What I mean by AI engineer
I'm not selling machine-learning research or model training from scratch. I'm an AI-augmented software engineer — someone who embeds LLM tools into how mobile and web products get built, reviewed, and released.
On production mobile and web I use Cursor, Claude, and Copilot across UI, PHP backends, and solo full-stack delivery. SolverIQ is a live example — AI roadmaps plus mentor booking. This portfolio is Next.js/React built the same way: AI drafts, I review, then it ships.
AI-assisted development workflow
The workflow is repeatable: read the codebase first, write short enforceable rules, use skills for recurring tasks, prompt with real file references, keep diffs small, and review AI output like a senior dev before merge.
I've written about this on the blog — React Native from zero with rules and skills, building a tiny PHP backend skeleton with AI, and AI web developer vs traditional web dev for hiring. Same discipline whether the stack is new to me or ten years deep.
What I can set up for your team or repo
Cursor or Claude project rules — stack constraints, naming, security bans, keep diffs small. Skills and playbooks for screens, API routes, service pages, backend endpoints. Prompts that reference your actual files, not generic tutorial fluff.
Review checklists for AI-generated code: SQL injection, secrets in repo, weak auth, wrong dependencies, navigation drift, and patterns that don't match your codebase.
LLM integration in products
When the product itself uses AI — chat flows, roadmap generation, mentor matching, content suggestions — I work on the integration layer: API calls, prompt boundaries, error handling, rate limits, and keeping user data out of logs.
SolverIQ is the public example: AI analysis plus human mentorship on a live platform. I know the difference between demo-quality AI and something you'd charge for.
Production sign-off — non-negotiable
AI makes drafts faster. It does not remove accountability for security, store guidelines, backend contracts, or regressions. Fast drafts, careful release — that's the service, not unlimited generated code.
If you want someone to paste Copilot output and walk away, I'm not the fit. If you want a senior builder who uses AI as leverage and still owns the release, we should talk.
FAQ
Questions I get asked
Should I hire an AI web developer or a traditional web developer?
Depends on scope. Traditional fits small fixed work or teams with a no-AI policy. AI-augmented fits faster page rollout, product AI features, or setting up Cursor on your repo — still with a senior review before production. I wrote a longer comparison on the blog.
Are you an ML engineer who trains models?
No — and I'll say that upfront. My AI engineering is applied: development workflows, repo guardrails, LLM API integration in products, and reviewing AI-generated code before production. For custom model training or data science, you'd need a different specialist.
Which AI tools do you use?
Cursor and Claude daily for code and architecture; Copilot in some environments. The tool matters less than the workflow — rules, scoped prompts, small diffs, and human review.
Can you help my team adopt Cursor or Claude safely?
Yes. Typical scope: audit how you work today, draft rules and skills for your stack, run a few features together with review, and leave you with templates you can reuse. Good for agencies and small teams who know AI helps but don't trust blind adoption.
Do you build AI features into mobile and web apps?
Yes — when the product needs them. Chat UIs, server-side LLM calls, admin tools around AI output, guardrails on prompts and responses. Mobile and PHP remain my production depth; AI is how I move faster and how modern products often differentiate.
How is this different from hiring a regular full-stack developer?
You get full-stack delivery plus deliberate AI workflow design — less thrash from huge unreviewed diffs, faster scaffolding, and honest limits on what AI should not touch without a human pass.
What's your main stack?
Day to day it's Framework7, Cordova, PHP, and MySQL — mobile, backend, and getting apps through Google Play and the App Store. I'm building React, React Native, and Next.js through actual projects (this portfolio included). I'll pick up new tools when there's a clear workflow and time to review before release.
How do you use AI without breaking production?
AI helps me draft UI, boilerplate, and scaffolding. It doesn't get the final say. I still walk through security, logic, integrations, and store requirements myself — same as I have for years of releases on live platforms. Fast drafts, careful ship.
Do you work with teams outside the Philippines?
Yes. I'm based in the Philippines and work with local teams and remote clients. English is fine, and I'm used to coordinating across time zones on production teams and direct collaboration.
Related
- Mobile app development
- Web development
- Full stack PHP
- AI web developer Philippines — pros, cons, and vs traditional web dev
- React Native with zero knowledge — how I use AI, rules, and skills to build from scratch
- Native PHP in 2026 — building a tiny backend skeleton with AI (and knowing when to stop)
- About me