Ahmet Muhsin Caliskan

Backend & Platform Engineer | Applied AI

Professional Summary

I build shared Spring Boot components for Vodafone Digital’s platform of 400+ microservices and co-developed DocForge, an LLM tool that turns code changes into service documentation. My five years of Java backend experience span telecom, banking and insurance.

Experience

Vodafone

Sep 2023 - Present

Backend Engineer | via Kafein Technology Solutions

  • Built modular Spring Boot 3/4 components for security, logging and caching, with a shared BOM and Gradle build conventions.
  • Reduced the infrastructure needed to run one production service by tuning JAXB2 marshalling. It used 140 GB less memory (a 70% reduction) and ran on 15 OpenShift pods, down from 40.
  • Helped build a Forge assistant that uses the Model Context Protocol (MCP) to retrieve framework documentation and coding conventions when engineers ask implementation questions. Contributed to the framework’s React.js dashboard.
  • Co-developed DocForge, an LLM tool that turns pull requests and API specifications into service documentation and release notes. The project received approval for production rollout.
  • Led migrations from Java 8 to Java 21 and from Hazelcast 3 to 5. Built a Spring AOP layer for distributed caching and maintained. Devleoped and designed architecture of eSIM migration journey, consent journey and usage info journey. etc APIs for Vodafone Yanımda. The planned upgrade to Spring Boot 4 includes Java 25 compatibility for Digital Framework and its microservices.

Türk Ekonomi Bankası (TEB)

Mar 2022 - Sep 2023

Full Stack Software Developer

Built Java and Spring Boot REST APIs for banking systems, wrote Oracle PL/SQL for core banking integrations, and developed internal applications with JSF and JSP.

Unico Insurance

Jul 2021 - Mar 2022

Full Stack Software Developer

Built Java EE and Spring Boot services that integrate insurance systems over REST and SOAP.

AI projects

Personal MCP server  Built an OAuth-protected server that lets an AI assistant answer questions about mobile usage and billing through read-only API tools.

Enterprise search (CASE)  Proposed a retrieval-augmented generation (RAG) design that uses embeddings and vector search to find relevant internal documents before an LLM generates an answer.

Technical skills

Applied AI
LLM integration, context engineering, MCP servers and tools, RAG architecture, fine tuning
Backend
Java, Spring, Spring Boot, Spring AOP, REST, SOAP, JUnit, Mockito, NodeJs
Platform
CI/CD (Jenkins), Gradle, OpenShift, Docker, Hazelcast, Git Actions
Databases
SQL, Oracle PL/SQL, PostgreSQL, MongoDB, Couchbase, S3

Education

B.Sc. in Computer Engineering

2015 - 2020

Eskişehir Osmangazi University