Skip to content
View KarlesP's full-sized avatar
🌴
On vacation
🌴
On vacation

Block or report KarlesP

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
KarlesP/README.md

πŸ‘‹ About me

I'm Panagiotis Karles, a Systems Cartographer with 10+ years of experience working at the intersection of software, data, security, and systems science.

I build systems that are practical, reliable, and easy to maintain β€” closing the gap between infrastructure, intelligence, and working product.

> profile --summary

Title        : Systems Cartographer
Experience   : 10+ years
Focus        : End-to-end product development with cloud and ML at the core
Learning     : Applied AI, LLM tooling, and developer experience patterns
Strengths    : Automation, product velocity, system reliability
Motto        : Push through the pain. Giving up hurts more.

πŸš€ What I do

βš™οΈ Engineering

I build maintainable software, APIs, CLI tooling, and automation pipelines β€” from rough idea to working system.

☁️ Cloud & Infrastructure

I work with cloud platforms (AWS / Azure) to design reliable, cost-aware infrastructure that supports fast shipping.

πŸ€– Machine Learning

I experiment with ML pipelines, LLM tooling, and applied AI for real product workflows β€” not demos for their own sake.

πŸ—οΈ Product Development

I care about the full product lifecycle: from identifying a problem to designing, building, and iterating β€” lean, fast, and with purpose.


🧰 Toolbox

Languages

Languages

Frameworks & platforms

Frameworks and platforms

Cloud & tools

Cloud and tools

πŸ† Featured work


πŸ¦€ cli-agent

Problem: No single-binary CLI agent that works with any OpenAI-compatible LLM for Linux pipeline assessments.

Stack: Rust, OpenAI-compatible APIs (Ollama, vLLM, OpenAI)

View repository β†’

⚑ tinyFastapi

Problem: A minimal, honest reference implementation for JWT authentication without heavy dependencies.

Stack: Python, FastAPI, TinyDB, JWT

View repository β†’


πŸ“Š GitHub snapshot

GitHub stats Top languages

GitHub activity graph

🧭 Engineering values

Readable beats clever.
Small systems beat over-engineered ones.
Good automation removes repeat pain.
Documentation is part of the product.
Security and reliability are not decorations.
Push through the pain. Giving up hurts more.

πŸ’‘ Approach

At some point we all touched a keyboard and turned a 0 into a 1. That moment is what I keep trying to replicate β€” but at the product level.

The goal is to close the loop between raw idea and working software: use cloud infrastructure to make it reliable, machine learning to make it smarter, and disciplined engineering to make it maintainable by someone other than yourself.

Build things that are useful, understandable, and worth maintaining.


🎲 Fun facts

  • 🧩 Escape rooms β€” nothing clears your head like a ticking clock and a locked door
  • πŸ”¬ Software experimentation β€” if it can be built, it should be tried
  • ✈️ Travelling β€” every new city is a different debugging session

🐍 Contribution snake

Contribution snake

🀝 Let's connect


Profile views

"Build things that are useful, understandable, and worth maintaining."



Footer

Popular repositories Loading

  1. edda-fastapi-demo edda-fastapi-demo Public

    This is an example fastAPI that showcases the use of JWT using a local tinyDB

    Python

  2. KarlesP KarlesP Public

  3. edda-android-demo edda-android-demo Public

    Kotlin

  4. cli-agent cli-agent Public

    Single-binary CLI coding agent for Linux pipeline assessments β€” Rust, OpenAI-compatible LLMs

    Rust

  5. exp-remote-sensing exp-remote-sensing Public

    Satellite image classification with Python β€” OBIA, supervised classification, and optional deep learning segmentation

  6. exp-data-analysis exp-data-analysis Public

    Reusable parallel data analysis patterns in Python and R β€” from concurrent.futures to Dask, from mclapply to furrr