Data Scientist

About Bitdefender

Bitdefender, one of the market-leading antivirus engines, is used by over 500+ million users worldwide. Our technology has been named Product of the Year by several internationally recognized testing organizations. As part of the Data Science team, you will develop and deploy models by leveraging the best tools and approaches to address business needs. You will experiment with different technologies and propose the best approaches to achieve your goals, working closely with the Data Analysis, Data Engineering, and Data Warehousing teams.

This position requires full-stack expertise in text-based generative AI, from prompt engineering, LLM orchestration and agents to training custom RAG components.

Responsibilities

• Engage in the full Data Science lifecycle: data collection and preprocessing, model training, deployment, and monitoring.
• Stay updated on innovations, tools, and techniques in large language models (LLMs) to enhance internal Data Science methodologies and make use of the best solutions for projects.
• Conduct ad-hoc and in-depth data-driven analysis to deliver clear insights and actionable recommendation.Compare models and collect data through experiment design and A/B testing.
• Handle multiple projects independently, ensuring well-documented processes and clear decision records.

Experience requirements

• Hands-on experience with large language model (LLM) tools and frameworks for prompt optimization, evaluation platforms, and orchestration (e.g. Haystack, LangChain, LlamaIndex), including RAGs and agentic behavior.
• Proficiency in classical machine learning (ML) algorithms as well as deep learning frameworks (e.g. PyTorch, TensorFlow) and architectures, particularly transformer-based models.
• Strong Python programming skills.
• Experience in SQL operations and automating data workflows with specialized orchestration tools (e.g. Airflow).
• Strong knowledge of statistics concepts (e.g. distributions, regression, statistical testing).
• Willingness and ability to learn new mathematical and/or technical methods.

Education

• Bachelor/Master’s degree in Computer Science, Statistics, Engineering or related fields.

Nice to have

• Hands-on experience with training and deploying deep learning models in production, including serving live predictions.
• Capable of translating intricate technical ideas into intuitive terms, with a clear focus on how and why they work.