Data Scientist, Deep Learning, NLP, LLMs
Pro Talent
full-time
Posted on:
Location Type: Remote
Location: Remote • 🇺🇦 Ukraine
Visit company websiteJob Level
Mid-LevelSenior
Tech Stack
AWSAzureCloudGoogle Cloud PlatformNumpyPandasPythonPyTorchScikit-LearnTensorflow
About the role
- Design, develop, and optimize ML models with a focus on deep learning, NLP, and LLM-based applications.
- Build scalable pipelines for training, fine-tuning, evaluation, and deployment of models.
- Work with frameworks such as PyTorch, TensorFlow, and Hugging Face Transformers.
- Fine-tune and adapt pre-trained LLMs (GPT, BERT, LLaMA, etc.) for domain-specific tasks.
- Develop solutions for text classification, summarization, embeddings, RAG, and conversational AI.
- Ensure model scalability, robustness, and low-latency performance in production environments.
- Collaborate with data engineers to prepare and optimize large-scale datasets.
- Implement MLOps practices (CI/CD, monitoring, retraining, governance).
- Participate in code reviews, documentation, and technical knowledge sharing.
Requirements
- 5+ years of experience in machine learning, with at least 3+ years focused on deep learning/NLP.
- Strong expertise in PyTorch or TensorFlow, and NLP frameworks (Hugging Face, spaCy, NLTK).
- Hands-on experience with LLMs (GPT, T5, LLaMA, Falcon, etc.), fine-tuning and prompt engineering.
- Proficiency in Python and libraries (NumPy, Pandas, Scikit-learn).
- Experience with MLOps tools (MLflow, Kubeflow, SageMaker, Vertex AI, or Azure ML).
- Strong understanding of transformer architectures, embeddings, and attention mechanisms.
- Familiarity with cloud platforms (AWS, Azure, GCP) for ML deployment.
- Excellent problem-solving and debugging skills.
Benefits
- Flexible work arrangements
- Professional development
Applicant Tracking System Keywords
Tip: use these terms in your resume and cover letter to boost ATS matches.
Hard skills
machine learningdeep learningnatural language processingLLMstext classificationsummarizationembeddingsMLOpsPythontransformer architectures
Soft skills
problem-solvingdebuggingcollaboration