02/16/2026 updated

**** ******** ****
100 % available

Software engineer, architect and developer - Python, Java, Elasticsearch, InfluxDB, Docker, AWS...

Zagreb, Croatia
Worldwide
Masters degree in software engineering
Zagreb, Croatia
Worldwide
Masters degree in software engineering

Profile attachments

HrvojeZeljkoCVv21.pdf

About me

I am an engineer with over 15 years of professional experience, working across a wide range of projects in complex and evolving environments. Focus is on delivering practical, well-considered solutions, with an emphasis on clarity, reliability, and long-term value rather than surface-level polish.

Java (Programming Language)Amazon Web ServicesAtlassian JiraDatabasesDevOpsDjangoElasticsearchPython (Programming Language)PostgreSQLMySQLNode.jsOracle ApplicationsShell ScriptSQLAlchemyReact.jsFlaskGrafanaVue.jsGitFastAPIKubernetesInfluxDBAtlassian ToolsApache KafkaFront End Software DevelopmentDocker
Strong Python and Java (+ Spring Framework). Big experience with Docker, Kubernetes, FastAPI, Flask, Django, SQLAlchemy, DBs like PostgreSQL, MySQL and Oracle, TS database InfluxDB and supporting ingestion and monitoring tools Grafana and Chronograf, Kafka, Elasticsearch, shell scripting and other supporting tools. Frontend experience include Vuejs and React.
Designed and developed several big systems using mentioned technologies.
10+ years of team-work using Atlassian tools, GIT and others.
Microservices architecure expert.

Experience

Professional experience10+ years
Freelancer since2017

Languages

GermanBasic knowledgeEnglishFluentCroatianNative speaker

Project history

Backend Developer

Atis Systems GmbH

Internet & IT

50-250 team member

Datapir

Datapir is a application/platform which provides big data integration, analysis, and AI-powered decision-making for government agencies, intelligence communities, and large commercial enterprises.

This type of platforms are often called "business intelligence on steroids" are intended for client with big amount of data to be deep analyzed:
  1. military (SIGINT)
  2. law enforcement (OSINT and SIGINT)
  3. intelligence agencies
  4. healthcare systems
  5. corporations
My responsibility was builing Search and Analysis Backend over ingested data, building LLM based query system to translate human readable/writable queries into structural data fetching queries. Knowledge of Ontology definitions, Ontop and SparQL was used among others.

Technologies and tools used:
  1. Python, Docker, Kubernetes, Kafka, Elasticsearch, Redis, Memcached, PostgreSQL
  2. Microservice architecture, FastAPI, Ontop, SparQL, Ontology, ANX, Ollama
Companies for which this project was done:
  1. Atis Systems GmbH
  2. Voice Collect Gmbh

Solution Architect and Full-Stack Developer

Trovent Security GmbH

Internet & IT

10-50 team member

Analytics Engine

Analytics Engine is many-in-one big data platform which gives possibility to design, improve and automate data driven processes.

ANX (Analytics Engine) is abstract platform which means it can be used for anything data related, especially Big Data related:
  1. building complex no-code ETL data pipelines
  2. CEP (Complex Event Processing) on micrsecond level utilizing Esper
  3. RAG arhitecture automated pipeline providing secure AI
  4. scheduled automated business processes
  5. orchestrating ML training and prediction processes
  6. log management platform for shipping logs to central data storage
  7. microservice deployment platform
  8. anything you can think of data oriented
Role:
  1. Design the microservices architecture for big data pipeline data flow platform
  2. Backend development of core components using Python, FastAPI, SQLAlchemy, Kafka, Java, Docker, Kubernetes, Elasticsearch
  3. Participate in Frontend development using Vuejs
  4. Builing base processors using Java, Python and Kafka
  5. Design the UX wireframes and prepare basic UX design
https://shepherd-analytix.hr/62e67000-5191-4e0a-b8ba-7a8d0053d47f/hrvoje-zeljko/#anx

Full-Stack Developer

Trovent Security GmbH

Internet & IT

10-50 team member

Eagle MDR

What is EAGLE MDR? It is a Managed Detection and Response automated system which helps detect anomalies in computer networks using static, dynamic and AI approach.

It relies on set of static and dynamic rules to detect anomalies as well as automated ML models. It relies on the MITRE ATT&CK framework, among others, as the basis for implementing the most effective detection of security-relevant incidents in order to ensure broad coverage of possible attack scenarios, methods and techniques.

I was part of the team to build this project from scratch:
  1. Java and Python BE developer
  2. partially Frontend development
  3. log management system design (ingest and storage of various log sources)
  4. UX design
Technologies and tools used:
  1. Java, Spring, Docker, Elasticsearch, Logstash, Kibana, MySQL, Kafka
  2. Python, Nginx, Redis, Memcached, Shell scripting, Esper, Vuejs, Websockets
Customers using EAGLE MDR:
  1. Trovent Security GmbH

Solution Architect and Backend Developer

Trovent Security GmbH

Internet & IT

10-50 team member

Closed RAG architecture for internal AI/LLM projects

A Closed RAG (Retrieval-Augmented Generation) architecture is a controlled AI/LLM system designed to operate entirely within a secure, internal environment. In this setup, large language models generate responses exclusively using curated, organization-owned data retrieved from private knowledge sources such as internal documents, databases, and knowledge bases.

No external data sources or public APIs are queried, and no sensitive information leaves the system boundary. This approach ensures data confidentiality, regulatory compliance, and consistent output quality, while still benefiting from the flexibility and reasoning capabilities of modern LLMs. Closed RAG architectures are especially well suited for enterprise use cases where accuracy, traceability, and security are critical, such as internal knowledge management, decision support, and domain-specific automation.

Use cases build upon this architecture:
  1. ISMS (Information Security Management System) AI agent
  2. Read-back/Hear-back voice agent
  3. Company employee onboarding AI agent
  4. MITRE ATT&CK tactic rule AI
  5. others in progress...
Technologies and tools used:
  1. Python, Opensearch, Elasticsearch, Memcached, ANX, Ollama

Backend Developer

Ergo Group AG

Internet & IT

1000-5000 team member

CLM Central Logging Management System

Log management and data processing system for shipping logs from various applications and system to Elasticsearch central storage. Monitoring and management of stored logs

Technologies and tools used:
  1. Elasticsearch, Logstash, Kibana, NiFi, Python, Shell scripting, Kafka
Customers:
  1. ERGO GROUP AG

Solution Architect and Full-Stack Developer

Shepherd Analytix d.o.o.

Internet & IT

< 10 team member

Solar Analytix

Application for smart exploitation of home solar power plant using machine learning and AI. Monitoring all parameters of the operation of the solar power plant, the use of an external electricity distributor, and suggesting to the user how and at what time to use devices in the house using artificial intelligence.

Technologies and tools used:
  1. Python, Docker, InfluxDB, Grafana, Chronograf, Kibana, Vuejs, React
Customers:
  1. private and internal project

Solution creator/architect and developer

Trovent Security GmbH

Internet & IT

10-50 team member

Logstash Syntax Checker
  1. Description:
  2. Real-time Logstash (from Elastic group) syntax checker, decreasing the time of Logstash configuration file syntax check from minutes to milliseconds
  3. Role:
  4. Build tool using Java, Ruby, Javascript, Node.js, Vuejs

Backend Developer

Trovent Security GmbH

Internet & IT

10-50 team member

Vulnerability Scan Monitor

Vulnerability Scan Monitor is a network vulnerability monitor system using external vulnerability scanner APIs. It provides automatic generation of reports and alarming

Technologies and tools used:
  1. Python, Flask, Memcached
Customers:
  1. Trovent Security GmbH
  2. indirectly to various companies as Trovent's SaS

Backend Developer

Ergo Group AG

Internet & IT

1000-5000 team member

PENTESTING Support System

System for managing automatic pentesting procedures over network and all parts of computer system

Technologies and tools used:
  1. Python, Django, Docker, Shell scripting, AWS, React
Customers:
  1. ERGO GROUP AG

Backend Developer

E2 Security GmbH

Internet & IT

10-50 team member

KNOW YOUR NETWORK (KYN)

Network traffic analysis and real-time anomaly detection system

Technologies and tools used:
  1. Java, Spring, Docker, Elasticsearch, Kibana, Logstash
Customers:
  1. E-ITO GmbH
  2. E2 Security GmbH

Fullstack JAVA Developer

IN2 d.o.o.

Internet & IT

500-1000 team member

Invest2 Front Office:
  1. IN2 d.o.o.
  2. financial asset management app
  3. LINK

Fullstack JAVA Developer

IN2 d.o.o.

Internet & IT

500-1000 team member

Performance Attribution Manager - PAM:
  1. IN2 d.o.o.
  2. investment decisions analytics tool

Solution creator and architect, Backend developer, Frontend Developer

NOVA TV

Internet & IT

500-1000 team member

World Cup 2026 Draw Simulator

Personal home project ahead of draw for World Cup 2026 held in December 2025. It is a continuation of previous World Cup 2018 draw simlator project which was done for Croatia's largest TV company NOVA TV and their sports portal Gol.hr. Project for World Cup 2026 was private in-house project while simulator for World Cup 2018 was provided to general public by NOVA TV.

Technologies and tools used:
  1. Javascript, CSS, HTML
Customers:
  1. NOVA TV (gol.hr)
  2. World Cup 2026 draw simulator (demo)
  3. World Cup 2018 draw simulator (demo)

Portfolio

Contact form

Log in to get in touch

You need to be logged in to use the contact form.

Sign upLog in