Senior Data Scientist
icims
Job Description
In This Role, You Will:
AI/ML Engineering & Delivery
- Design, develop, and deploy production-grade AI/ML solutions: RAG pipelines, NLQ systems, agentic AI workflows, and conversational search applications.
- Build and fine-tune LLM-powered applications using LangChain, LangGraph, LangSmith, and Langfuse; implement monitoring and evaluation pipelines.
- Develop supervised, unsupervised, and deep learning models across text, numeric, and multimodal data; deliver rigorous model evaluation and documentation.
- Write production-quality Python with full test coverage; follow CI/CD, Git workflows, and containerisation best practices.
- Build data products and dashboards on Streamlit and Sigma; integrate ML models with LLM capabilities and REST APIs.
Technical Guidance & Collaboration
- Mentor Data Scientists at Levels 1 and 2 through code reviews, pair programming, and technical Q&A.
- Contribute to solution architecture discussions and flag technical risks early in the project lifecycle.
- Translate business requirements into analytical frameworks; present findings clearly to both technical and non-technical stakeholders.
- Maintain thorough documentation of methodologies, model decisions, and results to support team knowledge transfer.
Here's What You Need:
Required
- Bachelor's in Computer Science, Mathematics, Statistics, or related field; Master's preferred.
- 4–6 years of hands-on data science / ML engineering experience, with at least 1 year working on LLMbased applications (RAG, NLQ, agentic workflows).
- Proficiency in Python (production-grade), SQL/Snowflake, Databricks, and AWS services.
- Hands-on experience with LangChain, LangGraph, or equivalent agentic frameworks; strong understanding of prompt engineering patterns.
- Solid grounding in ML fundamentals: regression, classification, clustering, tree-based methods, NLP, and deep learning.
- Experience with MLOps: model versioning, CI/CD, Docker, automated testing, and model monitoring. Strong written and verbal communication skills; able to present complex results to non-technical audiences.
Nice to Have
- Experience with vector databases (OpenSearch, Pinecone, ChromaDB) and embedding-based retrieval. Familiarity with model fine-tuning, LoRA/QLoRA, or RLHF techniques.
- Exposure to React/Node.js for full-stack data applications; experience with A/B testing frameworks.
- Knowledge of Sigma, Tableau, or similar BI/visualisation tools.