Skip to content
View BrightProgrammer7's full-sized avatar
🌚
Human
🌚
Human

Block or report BrightProgrammer7

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. Markdown is supported. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
BrightProgrammer7/README.md

Abdelilah Akhmim

Abdelilah Akhmim

Abdelilah Akhmim

Co-founder & CTO at JABB  ·  Casablanca, Morocco 🇲🇦

Portfolio  ·  LinkedIn  ·  X  ·  Projects

8 first places twice in Dubai #4 on F6S profile views

AI systems that turn field data into decisions

I build AI agent systems that turn scattered field data into decisions companies can act on. Right now that means LLM and offensive security tooling, customer and employee experience at JABB, and before that industrial safety and environmental sensing. State engineer in AI and computer science, ENSA El Jadida.

Fourteen competition placings in two years, eight of them first place, twice sent to represent Morocco in Dubai.

Now — Co-founder & CTO at JABB. Built the full stack in under three months: the consumer app, the voice-of-customer dashboard, and the CX platform underneath both. We also built the AI model that dispatches targeted surveys and in-field missions to JABBers, so clients get real data from the markets they target instead of panel estimates. Alongside it, GenAI Consultant at AI Startup Studio, taking business problems to shipped AI systems, architecture through deployment.

🔐 Security work

I build offensive security tooling with AI, not just AI products.

Project What it is
Pentasio 2026 Automated penetration testing platform, private beta. An adaptive agent runs Nmap, Nuclei, Subfinder, FFUF and Nikto across recon, enumeration and exploitation, scores findings by CVSS and writes remediation guidance. Nothing active runs until the target passes DNS, file or meta-tag ownership verification and signs a scoped 90-day authorization.
LLMGuardian 2024 Security assessment suite for teams shipping LLM features. A battery of checks, reported as what failed rather than one number. Won the Morocco national qualifier of the Arab Cybersecurity Challenge and went to Expand North Star, Dubai.

Currently focused on agent security: scope enforcement and authorization boundaries for agents that execute real actions against real infrastructure.

🚀 Selected work

Project What it is
H.A.S.S.A.N. 2025 Voice storefront for blind and low-vision shoppers. Voice-auth sign-in, then browsing, checkout and delivery all by speech.
GreenBee 2024 AIoT beekeeping platform. In-hive sensors plus models reading what the readings mean for colony health. ~150,000 MAD in grants.
Agenix-Eye 2025 Multi-agent system for industrial safety. YOLO detects on the floor, a RAG layer supplies the site's own safety context, LangChain decides when to raise a spoken alert.
AlHakikaNews 2025 Fake-news detection, LLM + classical ML. 95.3% accuracy over 44,000+ articles.
MonuMate 2025 AI tourist companion for Marrakech. NFC, QR and image recognition open a monument's story; RAG on a fine-tuned Falcon 7B handles follow-ups.
IAllo 2025 AI call-bot over WhatsApp and phone. Twilio Voice Media Streams + OpenAI Realtime, bridging caller, human agent and backend in one live conversation.
Ni'ma 2024 Forest fire and illegal-activity detection. ESP-32 over LoRa, cross-checked against satellite imagery.

See the full portfolio  ·  Visit jabb.cx

🏆 Recognition

  • 8× first place, national competitions
  • 2× represented Morocco in Dubai — GITEX Global and Expand North Star
  • GreenBee ranked #4 AI startup worldwide on F6S, out of 2M+ listed
  • AfriCHI 2025 Design Competition finalist, pitched in Egypt
  • Winner: Arab Cybersecurity Challenge (MA qualifier), Arab FinTech Challenge (MA qualifier), FINNOVATE'25, Nexus Hackathon, Capgemini Techathon

highlights

Before JABB
Orange Platform and governance layer for humans and AI agents collaborating across enterprise workflows.
Sofrecom Maroc HERMIONE to microservices — Spring Boot, Angular, CI quality gates. Converted to a full-time offer.
Capgemini SysML.AI — UML and SysML from plain language, Gemma-7B fine-tuned on 10k+ PlantUML/Mermaid samples.
Zenika Co-lunch network app on GCP — Cloud Run, Firestore, React/TS, Spring Boot.
JESA JESA Digital Estimator, an Electron tool automating engineering-hour estimation.
3D Smart Factory MeshSegNet for 3D dental scan segmentation, served over FastAPI on AWS Lambda.

⚙️ Engineering signal

profile scan

stack

LangChain · RAG · MCP · agent orchestration · ESP-32 / LoRa · Twilio · OpenAI Realtime

contribution streak repos per language

most committed language when I commit

📈 A profile that moves

Most of what I ship now is in private repos, so the graph below undersells it.

contribution activity, jet

contribution activity

🤝 Open to

Consulting on AI systems — architecture through deployment — and hackathon teams. Happiest talking about agent orchestration, RAG that survives real data, and getting LLM features safe enough to ship.

Portfolio  ·  Writing  ·  LinkedIn  ·  X  ·  Instagram  ·  Medium  ·  akhmim.abdelilah@gmail.com

Built in public from Casablanca.

Pinned Loading

  1. clickup-skills clickup-skills Public

    Agent Skills for ClickUp over its REST API — install via the skills CLI, compatible with Claude Code, Gemini CLI, Cursor.

    Shell 1

  2. cmd-toolkit cmd-toolkit Public

    Windows .bat scripts for the daily grind — backups, temp cleanup, task killing, env refresh, project scaffolding.

    Batchfile 1

  3. SpotifAI SpotifAI Public

    JavaScript 1