Skip to content
View arturio-amorim's full-sized avatar

Block or report arturio-amorim

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. Markdown is supported. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
arturio-amorim/README.md

Hi, I'm Arturio Amorim 👋

AI / LLM Engineer • ex–Senior Full Stack Engineer (10+ yrs) • Remote 🌎

I build production Generative AI — RAG systems, LLM agents, and AI-powered products — on top of a decade of shipping real software (Python/FastAPI, AWS, React). I care about systems that are evaluated, reliable, and actually deployed, not just demos.

🔭 Focus: Retrieval-Augmented Generation (RAG) · LLM agents & tool use · LLM evaluation/observability 🛠️ Stack: Python · FastAPI · LangChain/LangGraph · OpenAI/Anthropic/Hugging Face · Vector DBs (pgvector, FAISS, Chroma) · AWS · Docker 🌱 Currently: building production GenAI across domains — RAG, agents/MCP, LLMOps, and applied AI (e-commerce, support, marketing, agribusiness) 📫 Reach me: arturio.amorim@hotmail.com · LinkedIn

🚀 Featured Projects

🧠 Core AI / LLM engineering

Project What it does Stack
financial-rag-eval RAG over financial docs with a RAGAS evaluation harness — most demos skip eval; this one measures it Python · FAISS · LangChain · Claude/OpenAI
finance-agent-mcp LLM agent with tool use + an MCP server (define tools once, expose to the agent and MCP clients) Python · MCP · Claude/OpenAI
llm-eval-observability LLMOps — tracing (latency / error rate) + an eval harness (heuristics + LLM-as-judge) Python · LLM-as-judge
lora-finetune-finance LoRA fine-tuning of a small open LLM (PEFT + TRL) with before/after comparison Python · PEFT · TRL · 🤗

🛠️ Applied AI by domain

Project What it does Stack
ai-ecommerce-integration E-commerce — auto SEO enrichment + semantic product search (pluggable Shopify/mock connectors) Python · LLM · semantic search
ai-support-copilot Customer support — triage (intent/sentiment/urgency) + KB retrieval + grounded replies Python · LLM · triage
ai-content-studio Marketing — content generation + a measurable SEO analyzer that ranks A/B variants Python · LLM · SEO

🌱 Agribusiness

Project What it does Stack
agro-crop-advisor Crop recommendation from soil + climate via a transparent scoring engine + LLM advice Python · LLM · AgriTech
agro-commodity-insights Commodity price analytics (SMA, volatility, trend) + a grounded LLM market report Python · LLM · time-series

More landing here as I build in public.

🧰 Tech

Python FastAPI LangChain Hugging Face PostgreSQL AWS Docker TypeScript React

GitHub stats

Popular repositories Loading

  1. arturio-amorim arturio-amorim Public

    AI / LLM Engineer — RAG · agents · LLMOps · ex-Senior Full Stack

  2. financial-rag-eval financial-rag-eval Public

    RAG over financial documents with a RAGAS evaluation harness (faithfulness, relevancy, context precision/recall). Python · FAISS · Claude/OpenAI.

    Python

  3. finance-agent-mcp finance-agent-mcp Public

    LLM agent with finance tools (tool use), exposed as both a CLI agent and an MCP server. Claude/OpenAI · FastMCP · fully unit-tested tools.

    Python

  4. llm-eval-observability llm-eval-observability Public

    LLMOps toolkit: tracing + evaluation (heuristics + LLM-as-judge) for LLM apps. Pure-Python core, runs with no API key.

    Python

  5. lora-finetune-finance lora-finetune-finance Public

    LoRA fine-tuning of a small open LLM into a concise finance assistant (PEFT + TRL). Trains on a free Colab GPU; before/after comparison.

    Python

  6. ai-ecommerce-integration ai-ecommerce-integration Public

    Connect an e-commerce store + add AI: automatic SEO enrichment and semantic product search. Pluggable connectors (mock + Shopify). Claude/OpenAI.

    Python