AI/ML Subject Matter Expert – Data Analytics & GenAI

We are looking for a highly experienced AI/ML Subject Matter Expert with strong Data Analytics and GenAI implementation expertise. The role requires a hands-on individual contributor who can design, build, and operationalize scalable analytics solutions, KPI computation frameworks, data validation pipelines, machine learning models, and governed GenAI capabilities over enterprise data.

The ideal candidate will bring deep expertise in Python, SQL, analytics engineering, machine learning, Retrieval-Augmented Generation, and Agentic AI frameworks, and should be comfortable working with complex business metrics, enterprise data sources, and open-source LLM technologies to deliver reliable, explainable, and production-ready analytical solutions.

Responsibilities:

Analytics Engineering & KPI Development:

• Design, develop, and maintain robust analytics code using Python, pandas, and NumPy to compute, validate, reconcile, and operationalize key business metrics including costing, margins, QBR metrics, and operational performance indicators.

• Build efficient data transformations, optimize performance through vectorization and memory management, and implement repeatable data pipelines with testing, logging, validation, and reconciliation controls.

SQL & Data Extraction:

• Develop advanced SQL queries to extract, transform, and shape data from enterprise systems and cloud data warehouse platforms.

• Work with complex joins, aggregations, window functions, query optimization, and governed metric definitions.

Generative AI / Ask-the-Data Prototype:

• Implement a governed GenAI prototype enabling users to ask questions over structured and semi-structured enterprise data.

• Use Llama-family or comparable open-source models through Ollama, llama.cpp, vLLM, or similar inference frameworks.

• Build Retrieval-Augmented Generation and Agentic AI pipelines across structured and semi-structured data.

• Design chunking, embedding, retrieval, and reranking approaches.

• Implement agentic workflows using CrewAI, LangGraph, AutoGen, LlamaIndex Agents, or equivalent production-ready agent orchestration frameworks.

• Produce structured responses such as tables, JSON, and drill-down-ready answers.

• Implement guardrails for grounded responses, citations, traceability to source data, and safe handling of sensitive fields.

• Support evaluation of GenAI and agentic outputs for accuracy, groundedness, reliability, latency, and operational usability.

Machine Learning & Advanced Analytics:

• Apply light-to-moderate machine learning techniques where appropriate, including anomaly detection, outlier identification, cost variance analysis, feed failure detection, simple forecasting, trend analysis, model evaluation, and error analysis.

Experimentation, Evaluation & Deployment:

• Create reproducible experimentation workflows, including test question sets for LLM evaluation, accuracy and groundedness checks, latency profiling, and performance tuning.

• Package deliverables for deployment using Docker, configuration management, and produce clear technical documentation, runbooks, and handover materials.

Education:

Relevant degree in Computer Science, Data Science, Engineering, Mathematics, Statistics, or a related field.

Technical Skills (Must haves):

• Expert-level Python skills, particularly with pandas and NumPy: data cleaning and transformation; joins, merges, aggregations, and windowed calculations; time-series data handling; performance optimization, profiling, and memory management.

• Strong SQL expertise, including complex joins, aggregates, window functions, and query tuning and optimization mindset.

• Solid understanding of statistics and machine learning fundamentals, including feature engineering, model evaluation metrics, overfitting and validation concepts, and scikit-learn or equivalent ML libraries.

• Practical GenAI implementation experience with Llama models or comparable open-source LLMs; Ollama or similar local inference tools; RAG and Agentic AI frameworks such as LangChain, LlamaIndex, LangGraph, CrewAI, AutoGen, or equivalent; and embeddings/vector stores such as FAISS, pgvector, Weaviate, or Pinecone.

• Strong engineering discipline, including unit testing and data testing, logging and error handling, Git-based development workflows, CI basics, Docker, and environment management.

Need to have (Can be bridged):

• Hands-on individual contributor experience driving solution development from prototype through deployment-ready deliverables.

• Ability to work with complex business metrics, enterprise data sources, and governed analytical solutions.

• Ability to collaborate with business and technology stakeholders and independently drive solution development.

Good to have (Not essential):

• Experience with Snowflake or comparable modern cloud data platforms.

• dbt experience, including modeling, testing, and documentation.

• Experience with enterprise semantic layers or governed metric definitions.

• Experience building lightweight APIs using FastAPI or similar frameworks.

• Familiarity with enterprise security concepts such as RBAC, data masking, sensitive data handling, and audit logging.

Preferred Qualifications (Optional):

Experience across the typical technology stack: Python, pandas, NumPy, SQL, scikit-learn, Jupyter, Git, Docker, FastAPI, LangChain, LlamaIndex, Ollama, Llama-family models, FAISS, pgvector, Weaviate, Pine cone, and Snowflake or equivalent cloud data warehouse.