Easy JobSenior Data Scientist with NLP
Stermedia · Worldwide · Full-time · Remote
You’ll be working on diverse machine learning projects for local and international companies as well as in academic research. This will involve different phases of the end-to-end delivery – direct contact with the client, business analysis of the problem, coming up with an appropriate solution, implementation and moving it to the production environment.
Requirements
Key Responsibilities
● Design and develop end-to-end agentic AI solutions using modern LLM frameworks
● Build and deploy Generative AI applications using Python
● Develop and manage AI agents, including tool integrations, memory management, and reasoning workflows
● Implement agent orchestration for multi-agent systems and complex task automation
● Apply advanced prompt engineering techniques to optimize model performance
● Establish agent observability frameworks (monitoring, tracing, logging, evaluation, guardrails)
● Deploy AI solutions using container-based architectures (Docker, Kubernetes)
● Ensure scalability, reliability, and security of AI systems in production
● Integrate AI agents with enterprise APIs, databases, and external systems
● Continuously evaluate model performance and implement optimization strategies
Required Skills & Qualifications:
● Strong proficiency in Python programming
● Hands-on experience with Generative AI / LLM frameworks (e.g., LangChain, LlamaIndex, Semantic Kernel, etc.)
● Experience building end-to-end agentic solutions
● Practical experience in agent development and orchestration
● Expertise in prompt engineering and LLM optimization
● Experience with container-based deployments (Docker, Kubernetes)
● Experience implementing agent observability (monitoring, tracing, evaluation frameworks)
● Knowledge of REST APIs and backend integration patterns
● Experience deploying AI applications in cloud environments (AWS / Azure / GCP)
● Understanding of RAG architectures, embeddings, vector databases
Preferred Qualifications:
● Experience with multi-agent collaboration