portfolio

Linux-first guide to ship a recruiter-ready AI GitHub portfolio from the terminal: scaffold a clean repo with Makefile targets, lock lightweight deps, train a scikit-learn baseline with tests, enforce quality via pre-commit, ruff/black and GitHub Actions, manage data with Git LFS/DVC, add a Gradio demo, and containerize with Podman so anyone can clone, run, and trust it in minutes.

Hands-on guide to a Linux-native AI portfolio: provision a reproducible workstation (apt/dnf/zypper), build Bash-first data pipelines scheduled with cron/systemd, run offline LLMs with llama.cpp, train/package a small model with Make and a CLI, benchmark with hyperfine, and ship as Podman/Docker containers with GitHub Actions CI—plus copy-paste commands and a clean repo structure.

Show Linux-savvy hiring managers you can wire AI into real shell workflows with four Bash projects: a terminal chat assistant, AI commit message generator, log triage summarizer, and OCR-to-summary pipeline. The guide covers curl/jq-based OpenAI-compatible APIs, env vars, installs, robust scripting and security tips, and packaging your tools into a polished repo with demos to boost your portfolio.