I'm an AI/ML engineer with 6+ years of experience developing production AI systems across computer vision, NLP, and enterprise applications. My focus is on building robust, scalable solutions that deliver real functionality without unnecessary complexity.I approach AI development systematically - understanding the full product lifecycle requirements, architecting for maintainability, and ensuring solutions are built on solid engineering principles.聽
My experience spans from implementing CNN architectures for satellite imagery analysis to developing LLM-based agents for enterprise IAM systems, always with an emphasis on production readiness and long-term viability. My technical expertise includes integrating Agent systems, building RAG systems with vector databases, developing computer vision pipelines, fine-tuning and optimizing LLMs, and architecting distributed ML systems.聽
I work across the full ML stack, from low-level algorithm optimization to system architecture design, ensuring each component serves the overall product goals.Throughout my career, I've worked with government institutions, startups, and enterprise clients, delivering solutions that range from proof-of-concept implementations to production-grade systems. I combine deep technical knowledge with the ability to understand and translate business needs and requirements into well-architected technical solutions.
Experience: 7 years
Yearly salary: $70,000
Hourly rate: $40
Nationality: 馃嚪馃嚫 Serbia
Residency: 馃嚪馃嚫 Serbia
Experience
AI Developer
Spectre AI 2024 - 2025
Developed internal AI agent systems with advanced capabilities for user-oriented assistance with cryptocurrency market analysis. - Architected multi-level agent implementation using LangGraph - Implemented memory systems and artifact management for persistent agent states - Built LLM Agent utility systems for dynamic context management, artifact flows, and automatic smart history compression - Integrated toolsets for crypto market data analysis and statistics - Designed and created LSTM based models and pipelines for price and volatility predictions utilizing historic data Developed internal AI agent systems with advanced capabilities for user-oriented assistance with cryptocurrency market analysis. - Architected multi-level agent implementation using LangGraph - Implemented memory systems and artifact management for persistent agent states - Built LLM Agent utility systems for dynamic context management, artifact flows, and automatic smart history compression - Integrated toolsets for crypto market data analysis and statistics - Designed and created LSTM based models and pipelines for price and volatility predictions utilizing historic data Skills: Large Language Models (LLM) 路 LangChain 路 Vector Databases 路 Application Programming Interfaces (API) 路 Long Short-term Memory (LSTM)
AI Developer
ProTech 2024 - 2025
Focused on advanced LLM implementations and RAG system development. - Fine-tuned and optimized Llama based LLM models for domain-specific applications using LORA and Q-LORA techniques - Implemented LangChain powered RAG systems on AWS Bedrock with local and proprietary models - Developed custom tool-calling features for ad-hoc RAG flows - Implemented QDrant Vector databases for RAG dataflows, and coordinated data generation and injection procedures, and management of vector database contents. Focused on advanced LLM implementations and RAG system development. - Fine-tuned and optimized Llama based LLM models for domain-specific applications using LORA and Q-LORA techniques - Implemented LangChain powered RAG systems on AWS Bedrock with local and proprietary models - Developed custom tool-calling features for ad-hoc RAG flows - Implemented QDrant Vector databases for RAG dataflows, and coordinated data generation and injection procedures, and management of vector database contents. Skills: Large Language Models (LLM) 路 Artificial Intelligence (AI) 路 Machine Learning 路 Retrieval Augmented Generation (RAG) 路 Vector Databases 路 LangChain 路 Amazon Web Services (AWS) 路 LLaMA 路 Anthropic Claude
AI Consultant
Serbian govt. Office for Information Technologies and eGovernment (employed through UNDP-Serbia) 2023 - 2023
Served as AI consultant for the Government of Serbia through UNDP, assisting adoption of AI solutions across government institutions. - Assisted coordination between stakeholders and development teams for AI tool adoption - Provided technical insights on AI implementation across multiple departments Served as AI consultant for the Government of Serbia through UNDP, assisting adoption of AI solutions across government institutions. - Assisted coordination between stakeholders and development teams for AI tool adoption - Provided technical insights on AI implementation across multiple departments Skills: Python 路 Artificial Intelligence (AI) 路 Large Language Models (LLM) 路 Deep Learning 路 Data Science 路 Machine Learning 路 Teamwork 路 Natural Language Processing (NLP) 路 MLOps
Teaching assistant
Faculty of Technical Sciences in Novi Sad 2022 - 2023
Teaching assistant for graduate-level course "Methods for exploration and exploitation of data" focusing on machine learning and CNN architectures. - Organized and conducted practical exercises for graduate students - Developed course materials focusing on CNN implementations - Evaluated and graded student projects - Mentored students in machine learning oriented semestral projects Teaching assistant for graduate-level course "Methods for exploration and exploitation of data" focusing on machine learning and CNN architectures. - Organized and conducted practical exercises for graduate students - Developed course materials focusing on CNN implementations - Evaluated and graded student projects - Mentored students in machine learning oriented semestral projects Skills: Algorithms 路 Python 路 Artificial Intelligence (AI) 路 Keras 路 Convolutional Neural Networks (CNN) 路 Image Processing 路 Deep Learning 路 Data Science 路 Computer Vision 路 Machine Learning 路 Teamwork 路 TensorFlow
Co-Founder, Backend Engineer & System Architect
GG CyberSecurity Systems 2021 - 2024
Co-founded and led technical development of GGLock, a security-focused product. - Designed and implemented RESTful API architecture - Established optimized and secure data flow structures - Developed cryptographic algorithms for product security features - Led backend engineering and system architecture decisions Co-founded and led technical development of GGLock, a security-focused product. - Designed and implemented RESTful API architecture - Established optimized and secure data flow structures - Developed cryptographic algorithms for product security features - Led backend engineering and system architecture decisions Skills: Algorithms 路 Data Analysis 路 Python 路 Analytics 路 Artificial Intelligence (AI) 路 Keras 路 Image Processing 路 Deep Learning 路 Linux 路 Data Science 路 Databases 路 Machine Learning 路 Teamwork 路 TensorFlow 路 C++ 路 Natural Language Processing (NLP)
Machine Learning Engineer
Cinteraction 2020 - 2022
Developed computer vision and neural network solutions for real-time applications. - Built neural network pipelines for land usage classification using satellite imagery - Developed soil composition analysis models from satellite data - Implemented real-time human body pose estimation systems - Created facial expression detection and classification solutions Developed computer vision and neural network solutions for real-time applications. - Built neural network pipelines for land usage classification using satellite imagery - Developed soil composition analysis models from satellite data - Implemented real-time human body pose estimation systems - Created facial expression detection and classification solutions Skills: Algorithms 路 Python 路 Artificial Intelligence (AI) 路 Keras 路 Convolutional Neural Networks (CNN) 路 Image Processing 路 Deep Learning 路 Data Science 路 Computer Vision 路 Machine Learning 路 Teamwork 路 TensorFlow
Machine Learning Engineer
ProTech 2019 - 2020
Focused on computer vision projects and neural network optimization. - Developed computer vision solutions using Keras and TensorFlow - Optimized neural networks using TensorRT and OpenVINO - Implemented production ML pipelines - Worked with standard data science stack including NumPy, SciPy, OpenCV Focused on computer vision projects and neural network optimization. - Developed computer vision solutions using Keras and TensorFlow - Optimized neural networks using TensorRT and OpenVINO - Implemented production ML pipelines - Worked with standard data science stack including NumPy, SciPy, OpenCV Skills: TensorFlow 路 Keras 路 Algorithms 路 Data Science 路 Python 路 Computer Vision 路 Machine Learning 路 Teamwork 路 Artificial Intelligence (AI) 路 Convolutional Neural Networks (CNN) 路 Image Processing 路 Deep Learning
Skills
ai
backend
machine-learning
python
web3-py