Rafael Bernucci@rafael94072
1
Events attended
1
Submissions made
Brazil
1 year of experience
About me
My path has not been linear, and that is exactly what shapes what I build today. I began on the factory floor of the metalworking industry, where I learned the value of efficiency, precision, and operational discipline. Later, I moved into Financial Services at Santander, XP, and Safra, where I saw how sophisticated financial intelligence is concentrated at the top, while millions of informal microentrepreneurs in Brazil operate without data, fair credit, or proper guidance. That realization led me to transition into Software Engineering and Artificial Intelligence. Today, as Founder & CTO of MarIAna (Estrela do Brasil), I combine my accounting background from USP with modern cloud and AI infrastructure to build an AI-powered CFO system that transforms WhatsApp voice messages into auditable financial statements, bringing high-quality financial management to underserved entrepreneurs.
š¤ Latest Submissions
ProofRoute: Token-Efficient AI Agent
ProofRoute is a general-purpose, token-efficient AI agent for AMD Developer Hackathon ACT II Track 1. It handles all eight task categories through an accuracy-first hybrid architecture. Each raw prompt first reaches fail-closed deterministic solvers that answer only with a unique, reproducible proof. Remaining tasks are assessed by a fine-tuned, Q8-quantized FunctionGemma 270M. Its predicted intent and five scores feed a calibrated per-intent matrix policy that may select the embedded, text-only Gemma 4 E2B. Hash-pinned Wilson 90% and Nash/minimax guards may preserve or reject this decision, but never expand the validated local cohort. A second, independently fine-tuned FunctionGemma 270M performs tool planning. For supported inventory, recipe-cost, bounded-arithmetic, and ordering tasks, it produces allowlisted tool calls whose arguments, provenance, execution, and results are mechanically verified. A model-generated plan can never bypass deterministic proof. Unsupported tasks, invalid assessments, rejected plans, failed proofs, contract failures, or runtime errors safely fall through to an evaluator-authorized Fireworks model. Every remote request uses the injected FIREWORKS_BASE_URL and only models supplied through ALLOWED_MODELS. Before inference, ProofRoute extracts the raw prompt from the official JSON envelope. The Answer Contract Engine validates the free-form answer and reconstructs results.json deterministically. This reduced Fireworks input tokens by 51.9% in controlled testing while preserving byte-identical Kimi answers. The public linux/amd64 image embeds all three local model artifacts and requires no runtime downloads. Version v3.9.0-dual-functiongemma is 2.94 GB compressed and passed 758 automated tests, clean public-pull, manifest, OCI-label, official I/O, 4 GB RAM, 2 vCPU, disabled-networking, read-only/non-root, and 10 GB size gates. The scored v3.7.3 rollback achieved 84.2% accuracy using 4,198 Fireworks tokens.
13 Jul 2026
