DEV Community: Shashwat Srivastava

How I Built a Telegram Affiliate Bot That Posts Deals 24/7 — Full Python Source

Shashwat Srivastava — Fri, 24 Jul 2026 14:54:16 +0000

I wanted a Telegram channel that posts Amazon deals automatically. No manual work. Just set it and forget it.

So I built one.

What It Does

The system takes a JSON file with products and does everything:

Converts each product into 3 promotional formats (pain-fix, deal-alert, short-urgency)
Posts to Telegram every 4 hours via GitHub Actions CI/CD
Runs an interactive bot with 7 commands
Tracks every affiliate click per product via Netlify serverless
Detects new members and sends welcome DMs with referral rewards
Sends daily admin reports with member count and referral stats

All on autopilot. No VPS needed.

Tech Stack

Layer	Technology
Language	Python 3.12
Bot Framework	python-telegram-bot v20
MTProto Client	Telethon
CI/CD	GitHub Actions (every 4h)
Click Tracking	Netlify Edge Functions
Storage	JSON files + Netlify Blobs

Architecture

GitHub Actions (schedule: every 4h)
  ├── feeder.py     → Products → 3 content formats
  ├── poster.py     → Post to Telegram channel
  ├── referral.py   → Referral stats
  ├── reporting.py  → Daily admin report
  └── Netlify Deploy → Click tracker

Local Orchestrator (auto-start)
  ├── bot.py        → Interactive bot (7 commands)
  └── referral.py   → Join detection + welcome DMs

Why I Built It

I was running a Telegram deal channel manually. Finding products, writing posts, formatting links, tracking clicks. It took 1-2 hours daily and I kept missing posting schedules.

I wanted a system that:

Posts consistently (humans forget, machines don't)
Tracks every click to know what works
Grows itself via referrals
Costs nothing to run

How It Works

1. Product Feed (`product_home.json`)

{
  "products": [
    {
      "name": "Wipro 10W Smart RGBW LED Bulb",
      "price": "₹599",
      "mrp": "₹1799",
      "discount_percent": "67%",
      "link": "https://amzn.to/your-link",
      "category": "Home & Kitchen",
      "rating": "4.2"
    }
  ]
}

2. Content Generation (`feeder.py`)

Each product becomes 3 posts optimized for different reader psychology:

Pain-Fix: "Tired of dark rooms? Here's the fix."
Deal-Alert: "67% OFF — Wipro Smart LED Bulb"
Short-Urgency: "Limited stock. Grab it now."

3. Automated Posting (`poster.py`)

Posts to Telegram with inline buttons — "Buy on Amazon", "Share Deal", "Join Channel". Dedup logic prevents reposting the same product.

4. Click Tracking (Netlify)

Every click goes through go.js — a Netlify Edge Function that:

Records the product ID, timestamp, and referrer
Redirects to Amazon with your affiliate tag
Returns stats via stats.js

5. Referral System (`referral.py`)

When someone joins via a referral link:

They get a welcome DM with bot commands
The referrer earns points (tiered: 1x, 1.5x, 2x)
/referral shows progress bars toward next tier

The Bot Commands

Command	What it does
`/start`	Welcome + channel link
`/deal`	Random unposted deal
`/topdeal`	Best discount
`/search <kw>`	Search products
`/referral`	Get invite link + stats
`/about`	Bot info
`/contact`	Channel link

CI/CD Pipeline

GitHub Actions runs every 4 hours:

feeder.py → regenerate content from products
poster.py → post 1 deal to channel
referral.py --oneshot → log referral stats
reporting.py --daily → send admin report
Netlify deploy → update click tracker
Commit updated state back to repo

Zero infrastructure cost. GitHub Actions free tier is enough for months.

What I Learned

1. Consistency beats volume. Posting 1 deal every 4 hours outperformed 5 random posts daily. The audience knows when to expect content.

2. Hinglish content works. Mixing Hindi and English hooks ("Ghar ki deal chahiye?") increased engagement 3x compared to English-only posts.

3. Progress bars drive referrals. Showing users "3/5 friends toward 1.5x points" in /referral increased referral link generation significantly.

4. Serverless click tracking is cheap. Netlify's free tier handles thousands of redirects monthly. No database needed — Blob store is enough.

Getting Started

The full source code is available on Gumroad with a complete setup guide (SETUP.md).

Includes:

10 Python modules (1,500+ lines)
30 sample products with 90 pre-generated content items
Netlify click tracker with per-product analytics
GitHub Actions CI/CD workflow
Setup scripts for Windows (setup.bat) and Linux/Mac (setup.sh)
30-minute deploy guide

https://shaswat7.gumroad.com/l/tg-affiliate-system

Built with Python 3.12, python-telegram-bot v20, Telethon, GitHub Actions, and Netlify.

GitHub Profile README Generator — Free Tool, 3 Templates, 10 Seconds

Shashwat Srivastava — Thu, 25 Jun 2026 06:42:24 +0000

The Problem

Every developer knows the feeling: you click on someone's GitHub profile and see a clean, well-organized README with stats, pinned repos, and activity — then you look at yours. Empty.

You know you should make one, but:

Writing markdown from scratch is tedious
You don't know what to include
Keeping it updated feels like a chore

So I built something to fix that.

The Solution: AutoDev README Generator

AutoDev README Generator is a completely free tool that takes your GitHub username, fetches your public data, and generates a beautiful profile README in seconds.

No login. No signup. No payments. Just type your username and go.

https://autodev-kappa.vercel.app/readme-generator

Three Styles to Choose From

You get three different templates depending on your use case:

1. Professional (Default)

Clean, well-structured layout with stats bar, about me section, language badges, top repos, and recent activity.

# Your Name

<div align="center">
  <img src="https://img.shields.io/badge/Repos-30-blue?style=flat-square" />
  <img src="https://img.shields.io/badge/Stars-1k-yellow?style=flat-square" />
  <img src="https://img.shields.io/badge/Forks-100-orange?style=flat-square" />
</div>

## 👨‍💻 About Me

> Full-stack developer building scalable APIs

- 🌍 Location: India
- 🔗 Website: example.com
- 📊 GitHub: @username

2. Minimal

Simple and lightweight — just the essentials. Name, bio, stats bar, and language badges. Perfect if you want a clean, no-fluff profile.

3. Recruiter

Recruiter-ready format with tables, pinned repository cards, and a comprehensive stats table. Includes a score badge that recruiters can click to see your full analysis.

How It Works

Behind the scenes, it's pretty straightforward:

Enter your GitHub username
Choose a style (Professional, Minimal, or Recruiter)
Click Generate
Preview the README in browser
Copy or download — both free

The tool fetches:

Your profile data (name, bio, location, company)
Repository statistics (stars, forks, languages)
Recent public activity
Top repositories by stars

It then compiles everything into a ready-to-use markdown file.

What Makes It Different?

1. It's part of a bigger platform. This isn't just a standalone tool — it's part of AutoDev, which also includes:

A GitHub profile scoring system (score out of 100)
A developer leaderboard to compare with others
A shareable badge for your README
A local git automation agent (npx autodev-agent)

2. 100% free, no strings attached. Preview, copy, and download are all free. No account needed. No data stored. No paid tier.

3. Three distinct styles. Most generators give you one template. AutoDev gives you three — Professional, Minimal, and Recruiter — each optimized for a different audience.

Try It Now

Here's what you need to do:

Go to autodev-kappa.vercel.app/readme-generator
Type your GitHub username
Click Generate
Pick your favorite style
Copy or download the markdown
Paste it in your GitHub profile repo

That's it. Your profile goes from blank to beautiful in under 10 seconds.

What's Next?

I'm building this in public and would love feedback. Here's what's coming:

More README templates (dark theme, devops-focused, student-focused)
README auto-update (syncs with your GitHub activity)
Team/org profiles
Custom sections

If you have ideas, DM me on LinkedIn or open an issue on GitHub.

Built with Next.js and Sharp. Deployed on Vercel. Open source on GitHub.

DEV Community: Shashwat Srivastava

How I Built a Telegram Affiliate Bot That Posts Deals 24/7 — Full Python Source

What It Does

Tech Stack

Architecture

Why I Built It

How It Works

1. Product Feed (product_home.json)

2. Content Generation (feeder.py)

3. Automated Posting (poster.py)

4. Click Tracking (Netlify)

5. Referral System (referral.py)

The Bot Commands

CI/CD Pipeline

What I Learned

Getting Started

GitHub Profile README Generator — Free Tool, 3 Templates, 10 Seconds

The Problem

The Solution: AutoDev README Generator

Three Styles to Choose From

1. Professional (Default)

2. Minimal

3. Recruiter

How It Works

What Makes It Different?

Try It Now

What's Next?

1. Product Feed (`product_home.json`)

2. Content Generation (`feeder.py`)

3. Automated Posting (`poster.py`)

5. Referral System (`referral.py`)