← All guides
APP IDEA #37 · AI · LINKEDIN · SALES AUTOMATION

SBL.so — AI scrapes LinkedIn for leads, then sells for you while you sleep

Built by Shubham Kumar and Ayush Singh — both 19, Indian founders. SBL.so scrapes LinkedIn for high-intent leads (people who engaged with competitor posts, just changed jobs, just got funded), sends AI-personalised outreach, then an AI agent takes over every reply — handling objections, qualifying the lead, and booking the meeting. 50M+ messages processed.

50M+
LinkedIn messages processed on the platform
99.7%
Safety rate — no account bans
19
Age of both co-founders — Indian, no IIT
3x
Engagement reported by early adopters in 2 months
01 / HOW IT WORKS

What the app actually does

SBL.so's insight is that LinkedIn outreach dies the moment someone replies. Every other tool stops there — you're back to manual SDRs or losing the lead. SBL's AI takes over the conversation at reply — reads the context, handles objections, qualifies the prospect, and books the meeting. The entire sales conversation, automated.

1

User pastes a LinkedIn post URL or Sales Navigator search

SBL extracts everyone who liked, commented, or engaged with that post as leads — people who engaged with a competitor's post are already showing buying intent. Instant, targeted lead list.

2

AI researches each prospect and writes a personalised first message

Not a template. The AI reads the prospect's profile, recent activity, and company news — then writes a first touch that references something specific. Open rates are 3-5x higher than generic outreach.

3

AI sends the sequence and tracks all signals

Opens, profile visits, replies, connection accepts — the AI adapts the follow-up sequence based on what each prospect actually does. If they open but don't reply, the next message is different from if they didn't open at all.

4

AI takes over every conversation after the first reply

This is the core differentiator. When a prospect replies, SBL's AI — trained on 50M+ real sales conversations — responds contextually, handles objections, answers product questions, and books a meeting. You wake up to qualified calls in your calendar.

02 / INDIA POTENTIAL

Does this work as an India play

India has one of the world's largest B2B sales and business development workforces — millions of founders, agency owners, and sales professionals doing LinkedIn outreach manually every day. Most are still copying and pasting the same 3 templates. AI-powered outreach that actually reads the prospect and handles replies is almost completely absent from Indian B2B sales.

5M+
Indian B2B sales professionals and founders doing LinkedIn outreach manually every week — the exact audience that would pay for AI that handles their conversations for them.
WhatsApp
Indian B2B outreach is moving to WhatsApp alongside LinkedIn — an AI that manages both channels from one dashboard, switching context between a prospect's LinkedIn DM and WhatsApp message, is a complete India-first feature.
₹4,999/mo
Realistic India-priced entry tier — Indian agencies and bootstrapped B2B founders who close one deal a month using AI outreach will pay ₹4,999/month in a second. One client closes ROI in full.
Hindi outreach
AI-generated outreach messages in Hindi for regional Indian B2B markets — most SaaS tools in this category only write in English. Hindi outreach to SMB founders in tier-2 cities is an untapped channel.
03 / THE WEEKEND BUILD

Friday to Sunday, hour by hour

Scope to LinkedIn lead scraping + AI first message for v1 — skip the full conversation AI for the first weekend. Prove the outreach works before automating the reply handling.

Friday
Evening · 3 hrs
7–8 PM Set up a Next.js project, Supabase for auth/database. Build the LinkedIn URL input — user pastes a LinkedIn post URL or Sales Navigator search URL.
8–9 PM Build the lead extraction pipeline — use a LinkedIn scraping API (Proxycurl or PhantomBuster API) to pull everyone who engaged with the pasted post URL: name, title, company, profile URL.
9–10 PM Build the prospect enrichment step — for each lead, fetch their LinkedIn profile data (current role, recent activity, company funding status) to personalise the outreach.
Saturday
Full day · 7 hrs
Morning Build the AI message generator — send prospect profile data to Gemini API, receive a personalised first connection message and follow-up sequence (3 messages, spaced 3-5 days apart). No templates — each message references something specific about the prospect.
Afternoon Build the campaign dashboard — user sees all leads, message status (sent/opened/replied/booked), and a unified inbox for all LinkedIn replies across all campaigns.
Evening Add the AI reply handler — when a prospect replies, the AI reads the conversation history and draft a contextual response handling common objections (not now, send more info, how does this work). User approves before sending in v1.
Sunday
5 hrs
Morning Add WhatsApp outreach alongside LinkedIn — same lead list, same AI personalisation, but messages sent via WhatsApp Business API for Indian prospects who are more reachable on WhatsApp than LinkedIn.
Afternoon Razorpay integration — Rs4,999/month Starter (1 LinkedIn account, 200 leads/month), Rs9,999/month Pro (3 accounts, unlimited leads, AI reply handling).
Evening Test with a real LinkedIn post URL, extract 20 leads, generate AI messages, review quality, send 5 real connection requests. Record the demo for your first post.
04 / APP STACK

What you're actually building with

Nx

Next.js 14

Frontend + API routes

Campaign dashboard, lead list, unified inbox, and message generator in one framework.

Pc

Proxycurl or PhantomBuster API

LinkedIn scraping

Extracts lead profiles and post engagement data without triggering LinkedIn's bot detection.

AI

Gemini API

Message personalisation

Generates personalised outreach messages for each prospect based on their profile and activity.

Sb

Supabase

Auth + database

Stores campaigns, leads, message sequences, and reply conversation history.

Wa

WhatsApp Business API

WhatsApp outreach

Sends the same personalised sequence via WhatsApp for Indian prospects.

Rz

Razorpay

Payments

Rs4,999/month Starter and Rs9,999/month Pro. UPI-first.

05 / WHERE & HOW TO DEPLOY

Going live

Where: Vercel for the app, Supabase for the database. LinkedIn scraping runs through Proxycurl or PhantomBuster as a paid API — no hosting your own scraper. The main ongoing cost is scraping API calls per lead + LLM tokens per message. Model these per pricing tier before launching.

Push to GitHub, import into Vercel, add environment variables.
Sign up for Proxycurl (proxycurl.com) — they have a pay-per-call API for LinkedIn profile and post data.
Add env vars: SUPABASE_URL, SUPABASE_KEY, PROXYCURL_API_KEY, GEMINI_API_KEY, WHATSAPP_API_TOKEN, RAZORPAY_KEY.
Test LinkedIn scraping with 5 real post URLs before building the full pipeline — confirm data quality and completeness.
Point a custom domain at Vercel. The dashboard is where users spend their time — make it fast.
06 / MARKETING & REVENUE

Getting paying users

How to market it

  • Post the "I pasted one competitor's LinkedIn post URL — got 120 qualified leads who already showed interest — AI wrote personalised messages for all of them in 3 minutes" reel. Show the actual lead list and one AI message. This is the hook.
  • Run 3-5 reels/day targeting Indian B2B founders, agency owners, and sales leads — people already doing manual LinkedIn outreach who know exactly how painful it is.
  • The WhatsApp outreach angle is a dedicated India reel — "Most Indian B2B founders respond to WhatsApp faster than LinkedIn — here's how I reach them on both from one dashboard." This converts immediately.
  • Cold outreach the ironic way: use SBL itself to generate leads for SBL. Post about LinkedIn lead gen, scrape everyone who engages, send them the tool. First 50 clients acquired this way.
  • Target digital marketing agencies managing LinkedIn for clients — one agency account = 10x the revenue of a single founder account. Agency tier pricing is your growth lever.

Who pays, and why

  • Indian B2B SaaS founders doing LinkedIn outreach manually who want the personalisation of a great SDR without hiring one.
  • Indian digital marketing and lead generation agencies managing LinkedIn outreach for multiple clients who need to scale without adding headcount.
  • Indian freelancers and consultants using LinkedIn as their primary client acquisition channel — they need volume but can't afford to sound generic.
Scenario
Paying clients/mo
Revenue/mo
Slow start
30 × Rs4,999
Rs1,49,970/mo
Agency traction + reels
200 × Rs4,999
Rs9,99,800/mo
3-5 reels/day + agency tier
600 × Rs7,000 avg
Rs42,00,000/mo
07 / START BUILDING

Paste this into Claude or GPT

This prompt sets up the full build context so the AI scopes, plans, and starts coding the project with you from message one.

BUILD_PROMPT.txt
I want to build an AI LinkedIn lead generation and outreach automation tool for the Indian B2B market, inspired by SBL.so (built by Shubham Kumar and Ayush Singh, both 19, Indian founders — 50M+ LinkedIn messages process
ed), scoped to ship a working version in a single weekend. Core product: 1. User pastes a LinkedIn post URL. The system scrapes everyone who engaged with that post (liked, commented) using Proxycurl API — these are high-intent leads who already showed interest in the topic. 2. For each lead, the system fetches their LinkedIn profile (current role, recent activity, company stage, funding if any) for personalisation context. 3. Gemini API generates a personalised first connection message and a 3-message follow-up sequence for each prospect — referencing something specific from their profile, not a template. 4. A campaign dashboard shows all leads, message status (sent/opened/replied/booked), and a unified inbox for all LinkedIn replies. 5. When a prospect replies, the AI reads the conversation and drafts a contextual response handling common objections. User approves before sending in v1 — full AI automation in Pro. 6. WhatsApp outreach: same lead list, same AI personalisation, sent via WhatsApp Business API for Indian prospects who respond faster on WhatsApp. 7. Pricing: Rs4,999/month Starter (1 LinkedIn account, 200 leads/month), Rs9,999/month Pro (3 accounts, unlimited leads, AI reply handling). Stack: Next.js 14 for dashboard and API routes, Proxycurl API for LinkedIn scraping, Gemini API for personalised message generation, Supabase for auth and database, WhatsApp Business API for WhatsApp outreach, Razorpay for UPI-first subscription billing. Deploy: Vercel. Help me step by step: 1. Build the LinkedIn post URL input and Proxycurl API call to extract engaged leads. 2. Build the profile enrichment step — fetch full LinkedIn profile data for each lead. 3. Build the Gemini API call that generates personalised outreach messages per lead. 4. Build the campaign dashboard with lead list and message status tracking. 5. Build the unified reply inbox. 6. Build the AI reply draft handler for prospect responses. 7. Add WhatsApp outreach via WhatsApp Business API. 8. Wire up Razorpay for subscription tiers. Keep explanations short and India-context aware. WhatsApp outreach is first-class for Indian B2B leads. Flag clearly that Proxycurl charges per API call — model the cost per lead before setting subscription pricing. If I get stuck, tell me to ask @buildwithkanhaa.

Start building this weekend

Send me a screenshot of what you ship — it might be the next reel.

DM @buildwithkanhaa →
Want personal help building this?

I personally work with builders on AI, online income & content creation. Run @buildwithkanhaa & @rawyapster.

Work with me →