Aymen
Jeddou.

Software & data engineering student. I built and run HiresHere, a live job site with a data pipeline and an AI classifier behind it.

Open for a PFE in data or AI, 2027
Engineering degree, Software & Data Engineering, Faculty of Sciences (FSB), 2024 to 2027. Arabic, French and English.
Aymen Jeddou

PFE intern, data or AI

Aymen Jeddou · 2027
Available
PythonSQLData pipelinesRAGEvaluation
I built and run HiresHere: a daily pipeline over about 540 companies' job feeds, with an AI classifier and a verdict for 243 countries.hireshere.app, live since October 2026

What I've built

HiresHere

Live Solo, since October 2026

A remote job site that tells you which jobs you can actually apply for from your country, and quotes the line in the posting that proves it.

  • A daily GitHub Actions pipeline reads the public job feeds of about 540 companies (Greenhouse, Lever, Ashby, SmartRecruiters, Workable, Recruitee, Teamtailor) plus Remote OK and Jobicy, and keeps each employer's posting date.
  • Gemini batch reads each remote or hybrid posting once and fills a typed Pydantic schema: who may apply, time zones, seniority, the evidence quote.
  • Plain Python rules turn that into a verdict for each of 243 countries at export time, so adding a country costs no AI calls.
  • Astro on Cloudflare Pages, job pages rendered by Pages Functions, D1 for sign-ups, daily email alerts and a Telegram bot. English, French and Arabic.

11,566 jobs listed · 597 companies · 243 countries · $1.85 to classify 4,563 postings

Counted on the live site, 11 October 2026How

Jobs, companies and levels come from the site's own search index after that day's run; the cost is Gemini's batch estimate for the 4,563 postings classified the same day.

Python · httpx · Pydantic · SQLite · Gemini Batch API · Astro · Cloudflare · GitHub Actions

Sabeel (RoadScan)

2026 · team of 4In progress

A citizen photographs a pothole, a YOLOv8n model grades its severity, two reports within 5 m merge into one, and a RAG agent drafts the repair order citing the actual law, routed to the municipality or the roads authority.

  • I own the RAG work-order agent and the work sheets it produces.
Chroma · YOLOv8n (ONNX) · FastAPI · PostGIS · Flutter · React
Sabeel's model detecting a pothole: confidence 0.97, severity 4 out of 5

FSB Nexus

Summer 2026 · IDEALAB

A multi-agent assistant for my faculty (FSB) that answers from the faculty's own documents and shows its sources. Hybrid retrieval over pgvector and Postgres full-text, fused with reciprocal rank fusion.

FastAPI · pgvector · Mistral · Next.js · Alembic · CI
FSB Nexus · landing page
FSB Nexus landing page: Pose ta question. Repars avec la source.

FSBridge

2026Live

Student portal for FSB: GPA and honours on the Tunisian 20-point scale, timetables, PDF documents, and a Gemini assistant served through Supabase Edge Functions.

React · TypeScript · Supabase · Gemini

Mobility & Environment ETL

2025 · private

An Airflow DAG over seven public datasets (crash reports, EV registrations, air traffic, GTFS transit, EPA facilities, car sales): extracted, staged, cleaned with pandas, quality-checked and loaded into Postgres.

Airflow · pandas · PostgreSQL · Docker Compose

AgroScan

2026

A Flutter app: photograph a leaf or an insect and get one of 40 North African crop pests identified by Gemini, with treatment advice in French, English or Arabic, and a scan history kept offline.

Flutter · Riverpod · Gemini · Drift (SQLite)
The citrus leafminer moth, one of the 40 pests AgroScan recognises

One of the 40 pests in the app's catalogue (citrus leafminer). Photo: Franck A., CIRAD.

Where I've worked

June to August 2026
IDEALAB

RAG Engineer & Scrum Master, intern

FSB Nexus, the academic assistant above, built for my faculty.

hit@5, retrieval0.50 → 0.90
MRR, ranking0.395 → 0.776
Measured with the evaluation harness I built for the teamHow

An annotated set and a 100-question set, scored on hit@5, MRR, recall@10, answer rate and wrong-answer rate. The jump came from finding a vector-index misconfiguration, then adding hybrid search (vector plus full-text, reciprocal rank fusion).

  • Built a multi-agent RAG pipeline (orientation, academic, administrative, learning) that answers with source citations.
  • Ran the backlog for a team of 4, integrated every branch, and shipped with CI (69 tests), Docker and Alembic migrations.
July to August 2025
OAK Visuals

Automation & Data Processing, intern

Automated multichannel customer journeys (email, SMS, push) in Dynamics 365 Customer Insights, and wired their data into central reporting with Power Automate and Microsoft Fabric.

Open source

I'm a new open source contributor. I started with the tools and data I rely on for HiresHere: reporting and fixing problems in company directories, job-board datasets and AI frameworks, and sharing what running them in production taught me.

What I work with

Data

SQL, PostgreSQL + pgvector, Airflow, pandas, SQLite, MongoDB, Supabase, data quality checks

AI

RAG, hybrid search, evaluation harnesses, Gemini, Mistral, sentence-transformers, scikit-learn, YOLOv8

Engineering

Python, TypeScript, FastAPI, Pydantic, Astro, React, Flutter, Docker, GitHub Actions, Cloudflare

Engineering degree, Software & Data EngineeringFSB, 2024 to 2027, 3rd year
Integrated preparatory cycleFSB, 2022 to 2024
DataCampAI Fundamentals · Data Analyst Associate

Get in touch