Skip to content
View mrgutierrezmario's full-sized avatar

Block or report mrgutierrezmario

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. 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
mrgutierrezmario/README.md

Mario Gutierrez

M.G. Network and Technology Solutions — I build practical software around real infrastructure: services that run on hardware I manage, with the auth, storage, backups, monitoring and runbooks done properly — not just the demo.

The three apps below run in production on a Mac mini at home, are open source, and have a documented way back up when they break. The fourth project is the Kubernetes staging platform built around them.

Projects

InsiderTrack — who is buying, and did it work?

Tracks the stock trades of U.S. Congress members, corporate insiders (SEC Form 4) and institutional investors (13F) from their legally required disclosures. Scores every ticker on who is buying, weights each member by how their past buys actually did against SPY, and snapshots every score so the hit-rate of each label at 30/60/90 days is a number, not a claim. Live site — no account needed.

Python · FastAPI · PostgreSQL · React + TypeScript · APScheduler · Docker · Tailscale Funnel

InsiderTrack MCP — ask it in plain English

A Model Context Protocol server that lets Claude (or any MCP client) question InsiderTrack: "Where are several insiders buying their own stock this month, did anyone in Congress buy the same names, and how good is that member's record?" — three tool calls, one answer no page on the site can give. Read-only by construction, token-gated, rate-limited, audit-logged, stateless; eleven tools, two resources, two prompts.

Python · mcp SDK · httpx · Docker

AI Lecture Notes — the one that started it

Records a lecture from the browser (desktop or phone), transcribes it live with Whisper, writes structured notes as the class runs, and answers questions about it afterwards from the transcript, notes and slides. Exports to PDF, Word, Markdown, MP3, or the user's own Google Drive. In daily use for a graduate course since March 2026.

Python · FastAPI · WebSockets · faster-whisper · PostgreSQL · MinIO · React · Docker · OAuth 2.0

homelab-gitops — staging that proves the backups

A Kubernetes (k3s) staging environment, managed with Argo CD, that runs copies of all three apps restored from the previous night's encrypted backups — with production credentials scrubbed from each copy. Every merge to main is built by CI, pushed to GHCR, and deployed to staging by Argo CD Image Updater committing the new digest to git, so git log is the deployment history; a weekly job restores both apps again and smoke-tests them. Built and verified in September 2026, then paused until it moves to its own machine — the write-up of why is in the repo. Production is watched meanwhile by Prometheus + Grafana on the Mac.

Kubernetes (k3s) · Argo CD · Helm · Sealed Secrets · GitHub Actions · GHCR · Prometheus · Grafana

How they're built

  • Operations first: nightly encrypted off-site backups with a scripted restore drill, uptime monitoring, log rotation, a one-page operator runbook per project (what to do weekly, monthly, when it's down, when data is lost).
  • Boring on purpose: pinned dependencies, Dependabot with CI-gated auto-merge, secret scanning, CodeQL, branch protection, semantic versions and changelogs.
  • AI where it earns its place: pluggable providers (local Ollama, Claude, Gemini, OpenAI) with automatic fallback, routed per job so free local models do the bulk work and cloud keys are spent only where they matter.
  • Developed with Claude Code as a pair-programming assistant; every feature came from real use and the problems it surfaced.

What I work with

Languages Python, TypeScript/JavaScript, SQL, Bash
Backend FastAPI, SQLAlchemy + Alembic, PostgreSQL, WebSockets, APScheduler
Frontend React, Vite, TypeScript
AI Anthropic Claude, Google Gemini, OpenAI, Ollama; MCP servers; Whisper
Infrastructure Docker & Compose, Kubernetes (k3s), Helm, Argo CD, Linux and macOS servers, Tailscale, Cloudflare Tunnel, S3/MinIO, GitHub Actions
Observability Prometheus, Grafana, UptimeRobot, health endpoints built for monitors
Practices migrations, encrypted backups and tested restores, OAuth integrations, rate limiting, runbooks

Contact

mgnetsolutions.com · mario.gutierrez@mgnetsolutions.com

Popular repositories Loading

  1. lecture-note-app lecture-note-app Public

    Self-hosted lecture capture platform: real-time speech-to-text (Whisper), incremental AI note generation, and question answering over transcripts and slides. FastAPI · React · PostgreSQL · Docker, …

    Python 1

  2. insidertrack insidertrack Public

    InsiderTrack — congressional, corporate-insider (Form 4) and 13F whale trades from the official filings, scored per ticker and checked against the market. FastAPI · React · PostgreSQL · Docker + Ta…

    Python 1

  3. insidertrack-mcp insidertrack-mcp Public

    MCP server for InsiderTrack: ask an AI assistant who in Congress and which insiders are buying, what a ticker scores, and whether it worked

    Python 1

  4. mrgutierrezmario mrgutierrezmario Public

  5. homelab-gitops homelab-gitops Public

    Shell