Augusto Bastos

AI Engineer Β· RAG Β· LLM Integration Β· SaaS Builder

I shipped Lucky Cat: a live multi-tenant SaaS for restaurants, built solo in 2 months. It runs live for a real restaurant in Limerick β€” real orders and daily use; the Stripe billing engine is live with the first live payment in review. I build RAG pipelines and multi-agent systems.

1live SaaS Β· real pilot client
2 moidea β†’ live SaaS, solo
386tests Β· open-source Wavr
RAG+ multi-agent pipelines

Featured Projects

A live SaaS with a real pilot client, a RAG sales demo, and open-source systems work. Built solo, front to back.

Lucky Cat restaurant SaaS platform
Flagship Β· Live in Production
1 live client Stripe in review Multi-tenant Built solo, 2 months

Lucky Cat: Restaurant SaaS Platform

Multi-tenant SaaS with a live Stripe billing engine (Stripe Connect in review), Supabase RLS and Cloudflare Workers. Live in a restaurant in Limerick, Ireland with real orders and daily staff use β€” two apps: Ownly (owner dashboard: shifts, payroll, stock, cash-up) and Tillr (customer ordering PWA with real-time tracking).

Supabase + RLS Cloudflare Workers Stripe Connect Claude API
PitchPilot logo
White-label SaaS Β· Live Demo

PitchPilot: RAG-powered Sales Platform

Built on a RAG pipeline (Gemini + Supabase pgvector): when a field rep hits an objection mid-pitch, semantic vector search surfaces the best context-aware counter in under a second. White-label β€” rebrandable for any industry in under a day. Live demo deployed as VoltLine for the energy sector.

Gemini RAG / pgvector Supabase Cloudflare Pages
Γ‰tica de Spinoza 2026 book cover
AI Orchestration Β· Case Study

Autonomous Book Pipeline: 7-Agent Case Study

A 7-agent Claude pipeline (research β†’ draft β†’ edit β†’ export) that produced an 84-page book (PDF + EPUB, ~21,000 words) with zero human drafting. Shown here as a write-up: the pipeline source is private (the book ships commercially), with a free public sample linked below.

Claude API Multi-agent Python PDF / EPUB
Wavr β€” radar mark with wordmark
Systems Engineering Β· Open Source

Wavr: Explainable Multi-Modal Home Sensing

Fuses WiFi CSI, network scan, camera pose and mmWave radar into one explainable per-room state (confidence = agreement Γ— strength) on a top-down radar over a floor plan you draw in-app. Privacy-first by construction: loopback-only, zero cloud, camera frames never stored, positions never written to disk. 386 tests, every sensor mock-tested without hardware. AGPL, built solo.

Python + FastAPI Sensor Fusion WiFi CSI / mmWave YOLO pose

Technical Skills

The stack behind the projects above.

Multi-agent Orchestration
Claude API & Prompt Engineering
RAG / Vector Search
Supabase + Postgres + RLS
Stripe Connect
Python Automation
Cloudflare Workers

About

Brazilian builder based in Limerick, Ireland. I learned Python, Supabase, Cloudflare and Claude orchestration by shipping Lucky Cat β€” a multi-tenant restaurant SaaS that went from zero to live production in under two months.

The Graphic Design degree shows in the UI: spacing, contrast and layout are deliberate. I work fast, document clearly, and don't leave clients dependent on me for basic maintenance. Currently pursuing the Claude Certified Architect (CCA) certification.

Location Limerick, Ireland
Languages Portuguese (Native) Β· English C2 Β· Spanish (Basic)
Availability Open to freelance (Upwork) and remote roles
Currently building Lucky Cat Β· PitchPilot Β· CCA Certification

Got a build in mind?

AI automation, LLM integration, Python scripting. Short builds or long engagements, both fine. I usually reply the same day.

Email me