8000 jmxt3 (João Machete) · GitHub
[go: up one dir, main page]

Skip to content
View jmxt3's full-sized avatar
💭
Building • No distractions
💭
Building • No distractions

Highlights

  • Pro

Block or report jmxt3

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.

Content in all repositories owned by your account will be closed.
Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. 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
jmxt3/README.md

🎯 Focus Areas

I specialize in building robust, scalable AI systems that go beyond simple API calls. My expertise lies in architecting sophisticated agentic workflows:

  • Agent Harness Engineering: Designing the underlying frameworks, memory management, and state-persisting systems that transform LLMs into reliable, long-running agents.
  • Agentic AI & Multi-Agent Systems: Architecting complex interactions where specialized agents collaborate autonomously on long-horizon tasks.
  • Context Engineering: Rigorously optimizing the information flow, retrieval strategies, and system prompts that govern agent reasoning and behavior.
  • "Vibe-Coding" to Production: Leveraging advanced LLMs for rapid, high-velocity prototyping and then hardening those prototypes into reliable, full-stack applications (Python, Rust, TypeScript).

🚀 Featured Projects

  • aStack - A disciplined engineering harness designed to reliably execute complex, long-horizon AI engineering loops (built for Google Antigravity).
  • jettOS.ai – An agentic operating system enabling natural, multimodal interaction (Speak to Talk, Speak to Edit, Speak to Ask).
  • Pixelwood.ai – A cinematic AI visual engine, built on specialized agent graphs, streamlining complex creative pipelines for production teams.

⚡ Technical Philosophy

I believe true velocity is achieved not just by coding faster, but by engineering smarter. I merge disciplined software engineering principles with the generative speed of modern LLMs to rapidly bridge the gap between concept and production. My focus is on building the sophisticated harnesses and resilient feedback loops required to deploy agentic AI reliably, moving beyond experimental prototypes into impactful, scalable systems.


📫 Connect With Me

X   GitHub   LinkedIn

Pinned Loading

  1. gitscape.ai gitscape.ai Public

    Scan any agent skill — authored or compiled — for prompt injection, secrets, and malicious execution before it touches your agent. Or compile the long tail on the fly.

    Python 32 3

  2. pdf-splitter pdf-splitter Public

    Break large PDFs into smaller, manageable mini PDF chunks.

    Python 1

  3. transformers transformers Public

    collection of examples and resources to help you get started with using the Hugging Face `transformers` library

    Python

  4. TensorFlow-Examples TensorFlow-Examples Public

    Forked from aymericdamien/TensorFlow-Examples

    TensorFlow Tutorial and Examples for beginners

    Jupyter Notebook

  5. adk_examples adk_examples Public

    Google Agent Development Kit (ADK) Examples

    Python 1

  6. awesome-python awesome-python Public

    Forked from vinta/awesome-python

    An opinionated list of awesome Python frameworks, libraries, software and resources.

    Python 1

0