Master’s Scholarship Fellow – Data Scientist, NLP, and Semantic Search Systems

This is a fully remote position, open to applicants in Brazil.

📋 Description

• Conduct text preprocessing and enhancement, which includes tasks such as cleaning, tokenization, lemmatization, noise elimination, and feature extraction.

• Generate document and query embeddings utilizing models like Sentence-BERT, OpenAI Ada, or comparable alternatives.

• Store and index embeddings in vector databases, including FAISS, ChromaDB, Qdrant, or Pinecone.

• Construct semantic search, information retrieval, and RAG application workflows.

• Assess the quality of retrieval and generated responses using metrics like recall@k and MRR.

• Tailor pretrained models in PyTorch or TensorFlow for tasks including classification, similarity assessments, and information extraction.

• Experiment with various architectures, such as transformers and basic neural networks.

• Maintain documentation for pipelines, technical choices, and experimental outcomes.

• Utilize Git for version control of code.

⛳️ Requirements

• Ongoing or completed Master’s degree.

• A degree in Computer Science, Computer/Software Engineering, Information Systems, Statistics, Applied Mathematics, Electrical Engineering, Data Science, or related fields within the Exact Sciences and Engineering disciplines.

• Proficient in Python programming, including typing, data manipulation using pandas/polars, and managing virtual environments.

• Hands-on experience with NLP, encompassing tokenization, stemming/lemmatization, stopword elimination, and vectorization methods like TF-IDF, word2vec, or embeddings.

• Practical understanding of embeddings, semantic search, and RAG methodologies.

• Familiarity with vector databases such as FAISS, ChromaDB, Pinecone, or Qdrant.

• Experience with at least one deep learning framework: PyTorch or TensorFlow.

• Adequate knowledge to load pretrained models and conduct basic fine-tuning.

• Comprehension of analytical pipelines and experimentation in data science as applied to NLP.

• Proficient in Git for version control.

• Availability to work 40 hours per week.

• Preferred qualifications: Experience with LLMs, prompt engineering, and response evaluation.

• Preferred qualifications: Familiarity with PostgreSQL, MongoDB, and Redis.

• Preferred qualifications: Experience on real-world academic or professional projects that involve documentation and testing.

🏝️ Benefits

• Scholarship Stipend: BRL 9,000.00.

• Remote work opportunities available.

• 40-hour workweek.

• Duration of 9 months.