Arthur Franca

There are engineers who build pipelines. And then there are those who build pipelines that scale with AI integrated from the design phase.

I do both.

My journey began in Manufacturing Engineering at UNICAMP. I worked at multinationals like Saint-Gobain (Demand Planning and S&OP), where I led automation projects that eliminated over 20 hours of manual weekly tasks.

I collaborated with the Technology Operations team at Scanntech (Commercial Intelligence and Pricing), leading projects with retail industry partners that delivered measurable results, including the recovery of R$1.4 million.

In Data Engineering for LATAM, I orchestrated ETL pipelines using Airflow and JupyterHub and managed data ingestion into BigQuery for analytical modeling, ensuring real-time data consistency and reliability across operations in five countries. Additionally, I developed cohort analyses and ML models (ARIMA, Prophet, and Random Forest with depth tuning) to predict potential customer churn and create actionable plans based on the findings.

Projects I have built:

🧠 ClearMind AI Ops — a clinical automation SaaS platform featuring LLM-structured intake, a hybrid architecture (local Llama + OpenRouter), and a human-in-the-loop review pipeline. End-to-end design; currently in production.

🎙️ Vocal-Baseline-Monitor — an acoustic feature extraction pipeline (using praat-parselmouth and openSMILE) with MDC thresholds for longitudinal intra-patient monitoring. Published on GitHub.

💤 Apneia-Monitor — a multi-source screening system (Garmin, SnoreLab, APIs) consolidated into a pipeline with automated notifications and a clinical dashboard.

📄 AI Job Radar — a SaaS platform featuring LLM-based CV parsing, job matching, and Stripe monetization. Built and currently in production.

I currently collaborate on global language model evaluation projects with Micro1 (Palo Alto, California), keeping me at the forefront of AI systems in production. I am neither solely a technical expert nor just a strategist. I can design solution architectures—covering data, pipelines, and AI integration—and translate them into what matters to the business. My guiding principles are scalable systems and decisions augmented—never replaced—by data and AI. For me, "human-in-the-loop" isn't just a checkbox; it is a fundamental design principle.

I am seeking roles where I can architect data solutions at scale, learn new technology stacks, and integrate AI/LLMs as a key architectural differentiator—within companies that view data as a strategic foundation.