Hello, World! I'm Ruan 👋
🔭 Currently exploring LLM-powered applications and automation
🌐 Check out my website at ruancomelli.com
I'm a software developer who loves to code, but there's more to me than just crafting bugs programming:
- 🇧🇷 Brazilian living in the beautiful Florianópolis
- 🎓 Mechanical Engineer by formation—Bachelor's and Master's from the Federal University of Santa Catarina (UFSC)
- 💻 Software developer by passion (self-taught almost everything I know)
- 🦖 Dinosaur enthusiast (what's your favorite one?)
- 🐕 Proud father to two adorable chihuahuas who think they're much bigger than they actually are
- 🏐 Beach volleyball player with more enthusiasm than skill
- 🎮 Last fell in love with Hollow Knight
Sourcery – Senior Software Engineer (2021 – 2025)
At Sourcery, I contributed to building intelligent tools for developers:
- 🤖 Designed and implemented LLM-powered systems for automated code review and bug fixing
- 🔄 Re-architected and maintained an event-driven GitHub/GitLab bot infrastructure supporting natural language comment commands
- 📝 Implemented automated PR titles and summaries; Mermaid diagram visualization; and security-focused reviews
- 📊 Built evaluation datasets and scripts for AI-generated patches, using LangSmith for testing and observability
- 🛠️ Improved our CLI, GitHub Actions, and IDE extensions to streamline workflows and increase adoption
- ⚡ Led codebase modernization by replacing Black and PyLint with Ruff, making linting and formatting near-instantaneous
- 🔍 Enhanced our static analysis engine and added linting and refactoring rules
Elint – Tech Lead (2021)
- 📄 Led a team in developing a proof-of-concept for automated PDF data extraction, successfully delivering a solution that was adopted by the client
- 🔎 Architected and implemented a custom OCR solution that outperformed state-of-the-art commercial tools, including Google Vision, by combining Tesseract with modern open-source models
- ☁️ Designed and implemented serverless document processing pipelines using infrastructure-as-code
UFSC – Master's in Mechanical Engineering (2019 – 2023)
- 🧠 Applied deep learning to a computer vision problem in mechanical engineering
- 🔥 Developed models to estimate heat flux in pool boiling experiments
- 📊 Implemented and optimized convolutional neural networks using Keras/TensorFlow and AutoKeras
- 🔝 Pushed the state of the art in pool boiling heat transfer estimation:
- Reduced prediction error by 73%-80%
- Optimized CNN architecture 73%-96% smaller than previous state-of-the-art
- Processed 200,000+ boiling images across four heating surfaces
- 📝 Published research: Thesis | Paper | Code Repository | Thesis Repository
- Python development (libraries, packages, testing)
- LLM integration and prompt engineering (Pydantic AI, LangChain, LangSmith, LangGraph)
- Code quality tooling and static analysis (Ruff, MyPy, Tree-Sitter)
- CI/CD with GitHub Actions
- Machine learning and computer vision
- Git workflows and version control
- Type-safe Python development
- Scientific software development and reproducibility
- LLM system evaluation and testing
- 🦀 Learning the Rust programming language
- 🌐 Practicing full-stack development
- 🐙 Contributing to open-source projects
- 🔧 Building more efficient developer tools
- Rust:
- Add
const
support for float rounding methods #141521: Addedconst
support for rounding methods (floor
,ceil
,trunc
,round
,round_ties_even
) across all float types (f16
,f32
,f64
,f128
) - my first contribution to the Rust project!
- Add
I occasionally share my knowledge and experiences through talks:
Event | Talk | Date | Links | Description |
---|---|---|---|---|
PyNE2023 | Construindo um assistente de código utilizando LLMs | 2023 |
📝 Slides
🎥 Recording |
In this talk, I shared how to build a code assistant using LLMs, with a focus on practical examples and tools. |
PyBR2022 | Refatoração: boas práticas e automação | 2022 |
📝 Slides
🎥 Recording |
In this talk, I shared some tips and tools for refactoring code, with a focus on automating code refactoring. |
Note to self: give more talks in English!
Here's what my development environment looks like:
- Editor/IDE: Cursor as my main IDE, with Helix for quick edits
- Terminal: Alacritty with Zellij as multiplexer
- Shell: NuShell as primary, with Bash as fallback
- OS: Ubuntu 24.04.2 LTS
- Theme: Dracula (created by a fellow Brazilian 🇧🇷)
Shell Enhancers | Starship • zoxide • Atuin • bat |
Dev Tools | Mise • Just • uv • Ruff |
Git Workflow | Graphite • Lazygit • Delta • Difftastic |
Code Quality | Ruff • Mypy • pre-commit • Sourcery |
Python Stack | Pytest • Pydantic • LangChain • Typer • Rich • MkDocs |
Data & ML | NumPy • Pandas • Pandera • TensorFlow • Keras • AutoKeras • SQLAlchemy |
Code Analysis | Tree-Sitter |
Knowledge | Obsidian |
Note that this only includes public contributions.
I'm always open to interesting conversations and collaboration opportunities. Feel free to reach out if you want to discuss tech, share ideas, or just say hi!
- 💼 Work opportunities in software development, AI, and developer tools
- 🌟 Contributing to open-source projects, especially in Python and Rust
- 🧩 Collaborating on innovative developer tools and LLM applications
- 🔄 Exchanging ideas about code quality, automation, and developer experience
If any of these align with your interests or needs, don't hesitate to reach out!