Globant - Senior Data Scientist - Generative AI/Retrieval Augmented Generation (7-10 yrs)

We are seeking a Senior Data Scientist to join our AI and Data Science team. This role is ideal for someone who thrives on delivering cutting-edge AI solutions and has hands-on experience with Generative AI (GenAI), Retrieval-Augmented Generation (RAG), and advanced Machine Learning applications. You will collaborate with cross-functional teams to build intelligent systems using LLMs, transformers, and robust ML engineering practices.

Key Responsibilities :

- Design and deploy RAG-based architectures, optimizing retrieval and reranking pipelines.

- Work with a variety of LLM-based embedding models (e.g., ADA, BGE) and indexing algorithms (Flat, HNSW).

- Apply and refine prompt engineering techniques and perform LLM parameter tuning.

- Build and deploy scalable AI services using Python, Flask/FastAPI, and Docker.

- Develop and optimize ML/NLP models for use cases including multi-class and multi-label classification.

- Work extensively with transformer architectures, attention mechanisms, and embedding models (Word2Vec, encoder-based).

- Utilize SageMaker, ECS, Lambda, and S3 for model training, hosting, and inference pipelines.

- Perform rigorous model evaluation using metrics like precision, recall, F1-score, and perform hyperparameter tuning.

- Collaborate with DevOps and data engineering teams to ensure smooth end-to-end model lifecycle management.

Required Skills :

- Hands-on experience with Retrieval-Augmented Generation (RAG).

- Deep understanding of retrieval, reranking, and indexing algorithms (Flat, HNSW).

- Experience with LLM embeddings (ADA, BGE, etc.).

- Experience with prompt engineering and LLM fine-tuning.

- Strong experience with object-oriented Python and type hinting.

- Proficiency with Flask or FastAPI for API development.

- Working knowledge of Docker for containerized environments.

- Experience with multi-class and multi-label NLP classification tasks.

- Proficiency with transformer architectures and attention mechanisms.

- Solid understanding of word and sentence embeddings.

- Knowledge of deep learning optimization and loss functions.

- Understanding of foundational ML algorithms : Linear/Logistic Regression, Random Forests, RNNs (LSTM/Bi-LSTM).

- Strong grasp of ML/NLP evaluation metrics and model selection techniques.

Preferred Qualifications :

- Experience deploying models into production environments.

- Exposure to tools like Hugging Face Transformers, LangChain, or Vector Databases (e.g., Pinecone, FAISS).

- Familiarity with CI/CD and MLOps best practices.