An outbound engine that finds B2B startups likely to need GTM automation, enriches them in Clay, verifies their founders, drafts a grounded first message and waits for my approval in a private console.
YC companies screened in one run
377
passed the source gates and moved on to scoring
23
founders verified from YC pages, 0 search credits
19
messages sent without my approval
0
Status
Running on demand on my laptop (n8n in Docker), with the review console on Vercel
Stack
n8n, Google Sheets, n8n Data Tables, Neon Postgres, Next.js, Vercel, Clay, Tavily, Groq, Gemini, Discord, YC directory, Hacker News API
A daily job-intelligence pipeline that merges hundreds of remote postings, filters them with config-driven rules and scores the survivors against my profile with an explainable formula.
active job sources (12 wired)
11
postings merged per full run
466 to 617
at most this many reach the LLM per run
15
infrastructure cost
$0
Status
Deployed on a GCP Always Free VM, scheduled daily at 07:00 Tunis time
Plain rules screen every account before any model or paid call: domain, team size, industry, timezone, seen before. Spend stays predictable, and the model only sees what passed.
Each model call goes through a chain of providers set in config. A rate limit or an outage moves the call to the next model, and every failure is logged per provider.
Every run writes its counters and error samples to an ops tab. Tuning is a config change with a revision number, one variable at a time, with the run that shows the effect.
I studied software engineering in Nabeul and spent my final six months at Wavess, a Berlin B2B SaaS startup, working on a platform that turns market signals into GTM actions for small B2B teams.
That internship taught me the part of GTM work nobody demos: most of the job is throwing away the wrong accounts, the wrong people and the wrong sentences before anything reaches a customer. Since then I have built two systems around that idea, both on free tiers, both running on my own data.
I write the decision down before I build the workflow. Every project here has its decision log, its telemetry and the bugs that changed it.
Education
National Engineering Degree in Software Engineering (Master's equivalent)
IT Business School (ITBS), Nabeul
2023 to 2026
Bachelor's degree in Information Technology
ISET Nabeul (Higher Institute of Technological Studies)
2020 to 2023
Languages
Arabic
Native
English
Professional working proficiency
French
Reading and listening
Contact
Open to remote B2B contract work and junior GTM engineering roles. My working day overlaps fully with EU business hours. Open to relocation.