Data Scientist

Join our Data & Analytics team to build and productionize ML/DL/NLP/GenAI solutions under the guidance of senior scientists. Youll work end-to-enddata prep, modeling, evaluation, deployment, and monitoringwhile learning best practices for quality, safety, and reliability.
What Youll Do Modeling & Research Train and evaluate models (classification, regression, text classification/NER, embeddings, LLM prompting/RAG basics). Perform error analysis and A/B tests; document results and iterate quickly. Data & Features Explore and prepare structured/semi-structured/unstructured text using Python/SQL; ensure data quality and reproducibility. Build reusable feature pipelines and prompt templates. Production & MLOps Package models as APIs/batch jobs with FastAPI/Flask and Docker; write unit/integration tests. Add monitoring for accuracy/latency/drift; maintain experiment logs (MLflow/W&B). Collaboration & Communication Work with product/engineering to define metrics and prioritize work; present findings to technical and non-technical audiences. Minimum Qualifications Bachelors/Masters in a quantitative field (CS, Data Science, Math, Stats, EE) or equivalent project/internship experience. Proficient in Python and SQL; strong grasp of statistics and experimentation. Hands-on ML with scikit-learn and introductory PyTorch/TensorFlow. Basic NLP (tokenization, embeddings, Transformers familiarity) and exposure to LangChain or LangGraph. Git, notebooks, and clear written/verbal communication. Preferred (Nice to Have) Projects/internships using RAG, vector databases (FAISS/Milvus/Pinecone), or LLM eval frameworks. Data processing at scale (pandas/Polars; Spark/PySpark basics), orchestration (Airflow/Prefect). Cloud familiarity (AWS/GCP/Azure), Docker, and simple API development (FastAPI). Participation in hackathons/Kaggle/OSS or research publications. Success Measures (First 90 Days) Ship at least one ML/NLP or GenAI feature to staging/production with tests, docs, and monitoring. Establish baseline metrics and an experiment log; close a set of scoped bugs/tech-debt items. Tools & Tech You May Use Python, SQL, scikit-learn, PyTorch/TensorFlow, Hugging Face, LangChain, LangGraph, vector stores, FastAPI, Git, Docker, MLflow/W&B, Airflow/Prefect, cloud services (AWS/GCP/Azure).