About
I'm an artificial intelligence engineer with expertise in building accessible, intelligent systems. I like coding useful tools and look out for the little details that elevate user experience.
Right now I'm working with AI agent frameworks and machine learning models for digital signal processing. I also lead the team at Webra, where we build fullstack websites and integrate AI where it actually makes sense.
I've worked at startups like Webra and Stamn, and at fintech companies like Urdaten. On the side, I built an AI vocal remover using deep learning audio processing. I'm planning to add support for more instruments when I get the time.
In my free time, you can usually find me studying (no, really) cloud and cybersecurity concepts. I also like to constantly modify my Linux configs or just try out different distros. When I'm not in front of a computer screen, I'm probably playing tennis or the guitar.
Experience
2025 — PRESENT Building web applications with Next.js and Supabase. Recent projects include a restaurant ordering system with Stripe integration, a biometric loyalty program using fingerprint auth, and a POS system with GPS tracking.
- Next.js
- Typescript
- Supabase
- Vercel
- Stripe
- Resend
- Cloudflare
FEB 2026 — APR 2026 Built 4 autonomous AI agents that run 24/7 for Twitter automation, brand monitoring, and trend analysis. Created custom TypeScript skills that chain web scraping, LLM reasoning, and automated posting.
- TypeScript
- OpenClaw
- Firecrawl
- OpenRouter
- Linux
2024 — 2025 Replaced OpenAI's API with self-hosted open-source models, cutting costs by 90%. Rebuilt the document extraction pipeline using vision-language models instead of OCR, reducing manual fixes by 80%. Built a document verification system using web scraping and LLM comparison.
- Python
- Azure AI Foundry
- OCR
- LLMs
- Computer Vision
- Linux
Projects
AI-Powered Chatbot
Built an intelligent conversational agent using transformer models and retrieval-augmented generation. The chatbot can understand context, maintain conversation history, and provide accurate responses based on a custom knowledge base.
- Python
- LangChain
- OpenAI
- FastAPI
- React
Computer Vision Pipeline
Developed a real-time object detection and tracking system for video streams. Implemented custom YOLO models with optimization for edge devices, achieving 60+ FPS on standard hardware.
- PyTorch
- OpenCV
- ONNX
- Docker
- Redis
NLP Text Classifier
Created a multi-label text classification system for content moderation. Fine-tuned BERT models on custom datasets and deployed with automated retraining pipelines.
- Transformers
- Hugging Face
- MLflow
- Kubernetes
Recommendation Engine
Designed and implemented a hybrid recommendation system combining collaborative filtering and content-based approaches. Scaled to handle millions of users with sub-100ms latency.
- TensorFlow
- Apache Spark
- Cassandra
- Airflow
Contact
I'm always interested in hearing about new opportunities or collaborations. Feel free to reach out!
Or email me directly at arturomaciaas@protonmail.com