DEV Community: ⛦ ROSHAN

ProjectCortex — AI Browser Co-Pilot that Goes Where You Go

⛦ ROSHAN — Wed, 22 Jul 2026 06:49:50 +0000

ProjectCortex stands out by putting an AI co-pilot directly inside your browser — no tabs, no copying/pasting, no context switching. Just highlight and It intelligently summarizes, fact-checks, defines, and answers questions about any web page with built-in smart fallback across Groq and OpenRouter. </>

Key Features:
AI Co-pilot Panel : A draggable glassmorphic UI that lives on any page, offering 1-click summarization, instant fact-checking, context-aware definitions, MCQ solving, and free-form Q&A with typewriter streaming.
Smart Selection Bubble : Highlight any text on any website and get an instant contextual action menu — summarize, define, or fact-check without lifting your hands from the mouse.

Dual AI Provider Architecture : Auto-routes between Groq and OpenRouter free tiers with multi-model fallback chains (Llama 3.3 → Gemma 2 → Mixtral → DeepSeek) — zero downtime, zero configuration.

Full SaaS Backend : JWT-based hardware-bound licensing, admin dashboard for license lifecycle management, version control, API logs, and user feedback — all powered by Next.js + Supabase + Prisma.
Supported Languages: Python, JavaScript/TypeScript, Go, Rust, Java, C++, Ruby, PHP, and more.

AI Providers: Groq (primary), OpenRouter (fallback), plus any OpenAI-compatible custom endpoint.

Licensing: Hardware-bound JWT license keys with 24-hour offline grace period, periodic auth verification, and full admin CRUD console.

Built with: Chrome Extensions MV3, Next.js 16, React 19, Supabase, Prisma, PostgreSQL, Groq API, OpenRouter API, WebGL fingerprinting, MutationObserver, Turbopack, and javascript-obfuscator.
License: MIT — open source, self-hostable, extensible.

AI-Powered Indian News Aggregator That Cross-References Stories...

⛦ ROSHAN — Sat, 18 Jul 2026 06:44:52 +0000

AI-powered Indian news aggregator that cross-references stories across multiple trusted sources before publishing. The whole pipeline runs on its own — fetch, verify, write, publish.

AI Verification Pipeline :-

Automated Fact-Checking: Every story is verified using Groq Llama 3.3 70B before it reaches your screen.
Multi-Source Cross-Reference: Stories must be confirmed by 2+ independent trusted sources or they are discarded.
Credibility Scoring: Every article receives a 0–100 credibility score based on source authority, detail richness, and writing tone
Neutral AI Writing: Verified facts are rewritten into unbiased, factual summaries — no sensationalism, no spin.

Fully Autonomous :-

Runs 24/7: Fetches news from RSS feeds and APIs every 30 minutes — no human editors needed.
SHA256 Deduplication: Duplicate stories are automatically filtered out using URL hashing.
Domain Blocklist: Known satire, spam, and fake-news domains are blocked at the pipeline level.
Structured Logging: Full observability with detailed pipeline logs to stdout.

Live - https://verifiedindian.vercel.app/
GitHUB - https://github.com/roshhellwett/projectsentinel.git

Tech Stack :-

Layer Technology
Frontend Next.js 15 + TypeScript + Tailwind CSS
Backend Python 3.11 + FastAPI
Database Supabase (PostgreSQL)
AI Verification Groq API (Llama 3.3 70B)
AI Writing Groq API (Llama 3.3 70B)
Hosting Vercel (frontend), Railway (worker)

AI-powered Project Architectures and Interactive Graph Models...

⛦ ROSHAN — Sat, 18 Jul 2026 06:35:31 +0000

Unlike generic templates, our integrated AI Architect synthesizes your exact codebase context. It reads your project structure and generates custom architectural explanations, module descriptions, and usage instructions with unmatched precision and speed..

Web UI Studio: A modern browser SPA with real-time markdown previewing, template switching, and AI co-pilot chat.
Terminal CLI: Fast, scriptable commands perfect for CI/CD pipelines, Git pre-commit hooks, and rapid terminal workflows.
Interactive TUI Suite: An intuitive terminal menu system that guides you through generation without remembering any syntax.

PyPi - https://pypi.org/project/projectreadmegen
GitHUB - https://github.com/roshhellwett/projectreadmegen.git

Languages: Python, JavaScript/TypeScript, Go, Rust, Java, C++, Ruby, PHP, and more.
Package Managers: pip, poetry, npm, yarn, pnpm, cargo, go modules.
Frameworks & Build Tools: React, Next.js, FastAPI, Django, Flask, Vite, Docker.
Licenses: MIT, Apache-2.0, GPL-3.0, BSD, or proprietary detections.

Built a Tool That Transforms Your Linux Audio in One Command

⛦ ROSHAN — Fri, 15 May 2026 13:10:48 +0000

If you use Linux and care about audio quality, you've probably been through this: hunting for EasyEffects presets, manually copying .json files into the right folder, Googling IRS convolution files, and still ending up with mediocre sound.

I was tired of it. So I built projectpulsewire — and it hit 23 GitHub stars in its first month.

What is projectpulsewire?

It's a Python CLI tool that gives you instant access to 47 premium EQ presets and 404 IRS (Impulse Response) files for EasyEffects on Linux — all installable in a single command.

No manual file copying. No config hunting. Just:

pip install projectpulsewire
python -m projectpulsewire start

And your Linux audio stack is transformed.

What's Inside

47 Curated EQ Presets

Organized into categories you'll actually use:

Bass — Punchy Everyday, Deep Sub, Clean Boost
Genre — HipHop, Classical, Electronic, Rock
Voice — Podcast-ready, Call clarity, Broadcast
Brand — Harman curve, Beats-style, Sony signature
Dynamics — Loudness normalization, Night mode

404 IRS Files

Dolby Headphone virtualization
DFX surround simulations
Creative room correction
Bass enhancement convolutions
Headphone crossfeed profiles

Auto Setup Included

Not sure if you have PipeWire or EasyEffects installed? Just run:

python -m projectpulsewire setup

It auto-detects and installs everything you need — PipeWire, EasyEffects, and all required plugins.

Full Command Reference

# Interactive menu (recommended for first-timers)
python -m projectpulsewire start

# List all available presets
python -m projectpulsewire list

# Install a specific preset
python -m projectpulsewire install "Bass - Punchy Everyday"

# List IRS files
python -m projectpulsewire list-irs

# Install an IRS file
python -m projectpulsewire install-irs "Dolby Headphone"

# View installed presets
python -m projectpulsewire installed

# Remove a preset
python -m projectpulsewire remove "Preset Name"

Why I Built This

Linux audio has always been powerful but inaccessible. PipeWire changed the game technically — but the tooling around it is still rough. Most people don't realize that EasyEffects can make their $30 headphones sound like $300 ones with the right preset.

I wanted to close that gap. projectpulsewire is my answer: a dead-simple CLI that brings professional-grade audio tuning to every Linux user, not just the ones willing to spend hours in audio forums.

The Numbers

23 stars in the first month
60 commits and growing
v2.0.3 released on PyPI
MIT licensed — fully open source
Works on Arch, Debian, Ubuntu, Fedora

Try It Now

pip install projectpulsewire

GitHub: github.com/roshhellwett/projectpulsewire

If this helped you or you find it useful, a star on GitHub goes a long way. And if you have preset suggestions, contributions are open and welcome.

Built as part of Zenith Open Source Projects — a collection of developer tools built with the philosophy that open source is the first step of development.