About the Role
We are looking for a hands-on AI Engineer to design, build, and maintain conversational AI agents in cloud environments. The role includes enabling interactions via voice, chat, email, or APIs, connecting AI to corporate systems and external models, and implementing evaluation and feedback mechanisms to ensure continuous improvement. Strong Python and Python ML skills are required, along with experience in agentic frameworks and cloud infrastructure.
Responsibilities
•Develop and launch scalable conversational AI systems that operate across multiple channels, including chat, voice, email, and APIs
•Architect microservices and asynchronous systems for AI applications, ensuring high availability and performance
•Use Python to implement orchestration and decision logic — the control layer that manages tools, workflows, and system interactions
•Optimize AI models through fine-tuning and LoRA integration, making data-driven decisions between different approaches (FT vs RAG)
•Establish prompt engineering libraries and define evaluation protocols for prompt performance
•Design monitoring and evaluation frameworks, including dashboards, metrics, and human-in-the-loop feedback loops to maintain model quality and trust
•Work with data specialists to design pipelines for collecting, cleaning, and structuring information for AI agents
•Design CI/CD pipelines, container standards, and cloud strategies for infrastructure migrations, cost optimization, and high availability
Requirements
•Strong background in Python, mainly for hooking up AI models, automating stuff, and handling orchestration logic
•1–2 years of hands-on experience with Generative AI and large language models (LLMs)
•Practical experience in machine learning engineering: data preparation, working with vector databases in the cloud, and training/optimizing models
•Strong backend and frontend architecture experience
•Deep understanding of ML evaluation, tuning, and prompt engineering
•Experience working with cloud platforms such as AWS, GCP, or Azure, including container orchestration and CI/CD pipelines
•Experience with agentic frameworks (LangChain, LangGraph) and multi-agent system design
Nice to Have
•Comfort with ML frameworks such as TensorFlow, PyTorch, or Scikit-learn
•Experience deploying to cloud AI platforms (Azure OpenAI, Amazon Bedrock, GCP Vertex AI) or on-premise solutions (e.g., vLLM)
•Experience with enterprise AI platforms such as AWS AgentCore or Databricks AgentBricks or Google Agents Space, or Azure AI Foundry
•Experience with observability and monitoring tools and frameworks
Conditions
•Diverse multicultural, multi-functional, and multilingual work environment
•Global scope, international projects in different business domains: life sciences and healthcare, retail, and distribution, software and hi-tech, etc.
•Outstanding career development opportunities with a transparent career path and a roadmap to accelerate your journey
•Great learning and development opportunities, including in-house professional training, career advisory and coaching, sponsored professional certifications
•Knowledge-sharing with colleagues from EPAM's global tech communities
•25 days holiday per annum
•5 days of sick leave without medical certification
•Private health insurance for employees and discount for private health insurance for family members
•Discount for Fit Pass program
•Comfortable modern offices in Belgrade and Novi Sad or remote work from any location in Serbia
•Regular corporate, social and wellbeing events (parties, sport events, advanced CSR programs etc.)
•The remote option applies only to candidates who will be working from any location in Serbia
