EdTech / training · Business Automation · Case study

Sprintzeal: Lead Sphere, an AI buying-signal engine that scanned 7,643 posts in one LinkedIn run.

Sales starts each day with people who publicly asked about a course, already scored and pushed to Salesforce. Nobody prospects by hand, and the pipeline reports its own failures.

  • ClientSprintzeal, Dubai · UAE
  • IndustryEdTech / training
  • Timeline5 weeks
  • Type of buildProduction pipeline
Project

What the project was about.

The challenge

Sales worked cold lists while real buyers were asking "which PMP course?" in public on LinkedIn and Reddit. The first n8n version failed silently and used up enrichment credits.

What we built

We built an AI intent-signal lead engine for Sprintzeal. Replaced 5 fragile n8n workflows with one async Python 3.11 package on a Render daily cron, with --resume checkpointing.

How it works

1. Schedule (Render cron): 06:00 UTC daily.
2. Scrape (Apify): LinkedIn + Reddit.
3. Score (Claude): Intent + fit.
4. Store (Supabase): Dedupe + enrich.
5. Sync (Salesforce): Qualified leads.

What was delivered

Replaced 5 fragile n8n workflows with one async Python 3.11 package on a Render daily cron, with --resume checkpointing
Apify scraping across 60+ certifications, with a stream-and-drop memory pattern that fixed out-of-memory crashes
Claude scores intent, course and seniority with defensive JSON parsing; suppression list and Supabase lead schema
Waterfall enrichment Prospeo → Apollo (leads with no data are skipped on later runs) → Salesforce push for 28 priority countries
Tools

What it was built with.

Each tool has one clear job in the system.

Python 3.11Core pipeline
asyncioRuns jobs in parallel
ApifyScrapes LinkedIn and Reddit
ClaudeAI scoring and understanding
SupabaseDatabase and login
ProspeoEmail enrichment
ApolloContact enrichment
SalesforceSales CRM
RenderDaily cron hosting
Timeline

How the 5 weeks went.

  1. Start

    Discovery

    We mapped the current process, the systems involved and where time or leads were lost.

    Agreed scope
  2. Build

    Build

    The workflows, integrations and logic were built and shown in short demos.

    Working system
  3. Before launch

    Test and train

    Real cases were run through the system and edge cases were fixed with the team.

    Tested system
  4. Week 5

    Launch and watch

    The system went live with alerts and logs, and was watched closely after launch.

    Live in production
Outcome

What changed.

Measured results from the project.

500Qualified leads per Reddit run
0 hManual prospecting / week
0Silent pipeline failures / mo
5 wksFrom kickoff to live

Qualified leads per Reddit run

0
500
BeforeAfter

Manual prospecting / week

20 h
0 h
BeforeAfter

Silent pipeline failures / mo

6
0
BeforeAfter
Book a call

Want a result like this?

Tell us which case is closest to what you need and you get a fixed-scope quote within 24 hours.

What you walk away with:

  • A mapped funnelWhere your leads arrive and where they are lost.
  • A recommended buildMatched to your channels and tools, like the case above.
  • A fixed-scope quote within 24 hoursYours to keep, whether we work together or not.
  1. 1Book a free call

    Pick a time that suits you. The call takes 30 minutes.

  2. 2We map it and quote

    We map your funnel and send a fixed-scope quote within 24 hours.

  3. 3We build and launch

    Once you approve, we build it, test it and launch it.

Pick a time that suits you

Choose a day and time for your free 30-minute call. It is booked straight into our calendar.

Your details are used only to book this call. No spam, ever.

Client Reviews

5.0 average rating on Upwork and Fiverr