Santiago Gómez Acero — Digital product engineer · AI
Bucaramanga, CO · GMT-5 · available
01 /
Six case studies: context, technical decisions and measurable results.
01 — 2025 — 2026
A platform that launches digital ad campaigns on Tonic, Meta and TikTok with AI-generated content — from hours to minutes.
🔒 Client project · private repo
02 — 2025
A multi-role, bilingual certification platform in production — with tests, e2e and continuous deployment.
03 — 2025 — 2026
An editorial platform for creators and agencies: per-fanpage AI agents, a unified calendar and publishing to Facebook, Instagram, TikTok and YouTube.
04 — 2026
A Financial OS for Colombian accountants: payroll, invoice data entry and the tax calendar automated with AI.
05 — 2025 — 2026
Conversational English practice with real-time bidirectional audio, for Latin American professionals.
🔒 Client project · private repo
06 — 2025 — 2026
A multi-country, API-first editorial CMS aimed at perfect Core Web Vitals.
02 /
Sites, tools and experiments that also tell the story.
Kanban task management with sprints, real-time analytics and a gamification layer with credits and leaderboards.
A World Cup 2026 prediction league: 104 matches, a global leaderboard and scoring via idempotent SQL triggers with anti-copy RLS.
Commercial lead-capture site that contributed to selling 4 of 16 lots in a country-side residential project.
Institutional site with its own design system and a narrative landing where hand-drawn illustrations morph into real photography (GSAP).
Institutional site for a rural association in Colombia: 400+ active users in the first months.
03 /
The stacks I work with every day.
Award-winning public speaker (ULIBRO 2019) and recognized developer at IngenioTIC 2022. I lead projects end to end: gathering requirements with non-technical stakeholders, communicating decisions clearly and delivering under pressure. English B2.2 — comfortable in international teams.
04 /
I don't just build AI products — I build with AI. Coding agents, a multi-model ecosystem and open-source models are part of my everyday toolbox.
Claude Code and Codex as pair programmers: I direct agents that write, test and refactor under my judgement. They multiply delivery speed without sacrificing quality — every line goes through my review.
claude-code · codex
Claude, GPT, Gemini: every task has its ideal model depending on reasoning, multimodality, context window and cost. Knowing how to choose is part of the craft.
claude · gpt · gemini
Llama, Whisper and local models via Ollama when the case demands data sovereignty or zero cost per token. AI does not always live behind a paid API.
llama · whisper · ollama
What I learn using these tools becomes features: generation pipelines, agents, RAG and automations that already live in the projects above.
pipelines · agentes · rag
05 /
2022
Recognized for a mobile transport app for university students: spotting a real mobility need and solving it with code.
2023 — presente
Lead-capture website that contributed to selling 4 of 16 available lots. Astro + PHP/MySQL lead backend.
2024 — 2025
AI training program by Colombia's Ministry of Technology.
2025 — presente
Institutional site for a rural association: 400+ active users in the first months, direct stakeholder coordination.
2025
Multi-role certification platform live in production with tests, e2e and CI/CD — automatic deploy on every push to main.
2025 — 2026
Client platforms: ad campaigns generated with multimodal AI and an English coach with real-time voice conversation.
2027
Expected graduation: May 2027, Universidad Autónoma de Bucaramanga.
06 /
Santiago Gómez Acero · Bucaramanga, CO
I'm Santiago, a full-stack developer from Bucaramanga, Colombia. Over the last few years I went from building institutional websites to designing and shipping full platforms with AI: multimodal content-generation pipelines, real-time conversational coaches and certification systems running in production with CI/CD.
I finish my Systems Engineering degree at UNAB in 2027 while building for real clients. I'm obsessed with getting things to production: with tests, with metrics and with real users. If anything you saw here could help your team or your project, let's talk.
07 /
Tell me what you need to build and I'll get back to you with a concrete proposal.