Achraf Tajani
CV
Achraf Tajani

Achraf Tajani

AT

Full Stack Developer · AI & Prompting · Software Architecture

// stack principal

Java
Java
Spring
Spring
TypeScript
TypeScript
React
React
Angular
Angular
PostgreSQL
PostgreSQL
Docker
Docker

// skills

Skills

Full Stack

End-to-end web application development, frontend and backend.

JavaSpring BootReactAngularTypeScriptNode.js

AI & Prompting

AI-assisted project design, prompt engineering, Claude Code methodology.

Claude CodePrompt EngineeringAgent SDKArchitecture IA

Software Architecture

MVC, hexagonal architecture, microservices, REST API, design patterns and clean code.

MVCHexagonalMicroservicesAPI RESTDesign PatternsClean Code

UI/UX

Custom design system, mobile-first responsive, functional micro-animations.

Tailwind CSSFramer MotionFigmaDesign System

// experience

Experience

EDF R&D
EDF R&Dnow

Senior Developer

Feb. 2026 → present

JNGV — Jumeau Numérique de Générateur de Vapeur

  • Corrective maintenance: bug fixing, Redmine support, CI/CD, testing, Sonar
  • Evolutive maintenance: new features, UI adaptation, non-regression testing
Java SpringAngularPostgreSQLApache TomcatSonarCI/CD
France Travail
France Travail

Senior Developer

Jul. 2023 → Dec. 2025
  • Internal data migration tracking application
  • Legacy code refactoring (Perl, Bash, Java)
  • Technical solutions design for migration team
Java Spring BootQuarkusAlpine JSPerlBashPL SQL
BNF
BNF

Senior Developer

Nov. 2020 → Jul. 2023
  • Feature development and maintenance
  • POC design and team mentoring
  • Code review and best practices improvement
Java Spring BootReactAngularHTML/CSS/JS
KEOLIS
KEOLIS

Junior Developer

Mar. 2020 → Oct. 2020
  • Feature development, bug fixes, code reviews
IonicAngularNode.jsHapi JS

// projects

Projects

localhost/keyper
2 screens
KeyPer

KeyPer

ReactViteSupabaseTypeScript
localhost/horizon
2 screens
Horizon

Horizon

mobile
React NativeExpoTypeScript
https://dispatchr.neotlantic.com
2 screens
Dispatchr

Dispatchr

ReactPythonFastAPILeafletDocker

// approche

Most people use AI as an assistant. I orchestrate it as a team.

Every project starts with architecture, not a prompt. My agents operate within a structured framework I designed — with defined roles, rigorous context coherence, and results that hold over time.

Design first

No generated code without prior architecture. Structure always precedes generation.

Controlled context

A proprietary context system that gives agents the rigor of a senior developer.

Orchestration, not execution

I lead. Agents produce. The result is solid, maintainable, deployable.

The method stays proprietary — the results speak for themselves.