Portrait of Ahmad Samim Hamza

Samim Hamza. A Full Stack AI Developer and Software Engineer shipping production systems.

Full Stack AI Developer, Software Engineer, and Master's student in Artificial Intelligence at the University of Jyväskylä, with 5+ years of experience building production systems across backend, full-stack, and AI domains.

At Konecranes, I designed and deployed enterprise RAG and agentic applications using AWS Bedrock, Bedrock AgentCore, LangChain, and LangGraph — turning messy internal documents into reliable answers through pipelines indexed from 100+ XML, PDF, and XLSX sources. I built the full stack (Next.js + FastAPI), evaluated three RAG variants, stored feedback in PostgreSQL for iteration, deployed on AWS Amplify/ECS/RDS, and documented the system for handover to the internal team. I work daily with AI-assisted coding tools such as Claude Code and GitHub Copilot.

Before this AI-focused chapter, I built ERP, logistics, and e-commerce platforms with Python, Node.js, TypeScript, React, Next.js, NestJS, and Laravel — including leading a team of eight engineers on Dara.af, an e-commerce platform with hundreds of product listings and daily customer orders. I'm equally comfortable on the frontend, designing backends and data layers, and deploying with Docker, AWS ECS, and Kubernetes.

Work Experience

  • Summer Trainee, Software Development Specialist (AI/ML) at Konecranes (05/2026 – Present): Built multiple production-ready RAG applications on AWS Bedrock using Agents and AgentCore, processing 100+ internal XML, PDF, and XLSX documents indexed in an S3-based vector store. Implemented RAG variants with AWS Bedrock, Bedrock AgentCore, LangChain, and LangGraph in a unified Next.js + FastAPI application with a shared conversational UI — developed primarily with Claude Code. Architected a multi-agent system with a supervisor agent orchestrating collaborator agents. Used PostgreSQL for user accounts, session data, and feedback storage to evaluate three RAG variants. Deployed on AWS Amplify, ECS, and RDS; collaborated with stakeholders, documented the solution, and handed it over to the internal team.
  • Senior Software Engineer at Ramaki Group of Companies (05/2024 – 07/2025): Led and coordinated a team of eight engineers on Dara.af, an e-commerce platform supporting hundreds of product listings and daily customer orders. Designed and optimized PostgreSQL schemas, built RESTful APIs, wrote unit, integration, and feature tests, and conducted code reviews while mentoring junior developers — using Laravel, React, and Next.js.
  • Software Engineer at Peace Global Logistics (09/2022 – 04/2024): Developed logistics and supply-chain modules using NestJS, React, and Next.js; designed PostgreSQL databases and optimized complex queries; implemented reusable UI components in React + Tailwind; and collaborated with cross-functional teams to deliver ERP features for 200+ daily users.
  • Instructor (Python & Machine Learning) at Kabul University (04/2022 – 10/2022): Taught Python programming and Machine Learning fundamentals to freshman and sophomore students, covering data analysis, model training and evaluation, and practical ML tooling.
  • Full Stack Developer at Smart Friqi (26/07/2021 – 20/09/2022): Built ERP modules using Laravel, Nuxt.js, and PostgreSQL; added real-time functionality with WebSockets; and developed dashboards and admin interfaces for business operations.
  • Web Developer at TwistedBits ICT (01/05/2021 – 15/07/2021): Built two management information systems using PHP, JavaScript, Laravel, and Vue.js, delivering projects under tight deadlines with a small team.

Education

  • Master’s Degree in Artificial Intelligence at the University of Jyväskylä (08/2025 – Present)
  • Bachelor of Computer Science (Software Engineering) from Kabul University (2017 – 2021)

Languages

  • English
  • Persian
  • Finnish

Top Skills

  • LangChain / LangGraph & RAG
  • AWS Bedrock (Agents, AgentCore)
  • Python, FastAPI & TypeScript/NestJS
  • Next.js, React & Laravel
  • PostgreSQL & Docker/Kubernetes
  • PyTorch & scikit-learn
  • Claude Code & GitHub Copilot