robhorvat

robhorvat / news-sentinel-mlops

Public

Production NLP MLOps news classifier with TF‑IDF baseline and PyTorch TextCNN, evaluation gates, FastAPI dashboard, drift checks, CI/CD, Docker, Kubernetes (minikube) and Prometheus metrics.

11
0
100% credibility
Found Feb 21, 2026 at 11 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
Python
AI Summary

News Sentinel is a demonstration project for building, evaluating, serving, and monitoring machine learning models that classify news articles into World, Sports, Business, or Sci/Tech categories.

How It Works

1
🔍 Discover News Sentinel

You stumble upon this handy project that sorts news headlines into categories like World, Sports, Business, or Sci/Tech using smart pattern recognition.

2
💻 Get it running on your machine

You follow easy steps to set everything up on your computer, like creating a quiet workspace for it to learn and work.

3
📚 Load sample news stories

You bring in ready-made examples of news articles so the sorter can practice on real-world headlines.

4
🧠 Train the sorters

You build a quick reliable sorter and a more advanced one, then compare their accuracy to pick the winner.

5
🚀 Open the control dashboard

Your web dashboard pops up in your browser, ready for action with charts and buttons.

6
🔮 Classify headlines live

You paste in news headlines, watch them get sorted instantly with confidence scores, and see performance stats update in real time.

7
📋 Create reports and summaries

You generate detailed proof reports and optional smart summaries explaining each classification.

✅ News monitoring system live

You now have a complete, watchful system for sorting news, tracking performance, and spotting issues, perfect for demos or daily use.

Sign up to see the full architecture

6 more

Sign Up Free

Star Growth

See how this repo grew from 11 to 11 stars Sign Up Free
Repurpose This Repo

Repurpose is a Pro feature

Generate ready-to-use prompts for X threads, LinkedIn posts, blog posts, YouTube scripts, and more -- with full repo context baked in.

Unlock Repurpose
AI-Generated Review

What is news-sentinel-mlops?

News Sentinel delivers a full Python MLOps pipeline for classifying news into World, Sports, Business, or Sci/Tech categories using a TF-IDF baseline classifier and PyTorch TextCNN. It tackles the jump from notebook experiments to nlp production by adding evaluation gates, drift checks, and quality baselines that block weak models. Users spin up a FastAPI service with predict/incident-summary endpoints, a live dashboard, Prometheus metrics, Docker images, Minikube Kubernetes deployments, and GitHub Actions CI/CD for github production deployment.

Why is it gaining traction?

Quality gates enforce production rules in nlp—TextCNN fails vs baseline right now, proving the system works. Drift checks, CI/CD pipelines, and a demo probe generate JSON/MD reports for proof-of-concept. The Makefile handles data prep, training, evaluation, and docker builds, while optional Gemini summaries fallback deterministically, making github production code feel battle-tested.

Who should use this?

MLOps engineers building nlp production classifiers needing baseline comparisons, drift checks, and evaluation gates. ML interviewees demoing end-to-end github production environment with FastAPI dashboard and Kubernetes manifests. Teams evaluating github production planning for CI/CD ML services with Docker observability.

Verdict

Grab it as a reference for nlp production despite 10 stars and 1.0% credibility score—polished Makefile workflows, solid docs, and CI tests make it mature for learning or forking into your github production release asset. Underperformant TextCNN is a feature, not a bug.

(198 words)

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.