Senior Data Scientist – Model AI & LLM Systems

Requirements:

1) Large Language Models (LLMs) & Agentic AI -

• Hands-on experience designing and productionizing LLM-based systems beyond chat

use cases.

• Strong expertise in prompt engineering for structured outputs (JSON, schemas, APIs,

models).

• Experience building multi-agent / agentic workflows (planner, reasoning, validation,

execution).

• Familiarity with tool-calling, function calling, and agent orchestration frameworks.

• Ability to ground LLMs using RAG, metadata stores, and domain knowledge.

2) Machine Learning & NLP Foundations -

• Strong grounding in NLP concepts (embeddings, similarity search, classification).

• Experience designing confidence scoring, ranking, and ambiguity detection mechanisms.

• Ability to evaluate and benchmark model outputs using custom metrics and test sets.

• Understanding of model bias, hallucination risks, and mitigation techniques.

3) Explainability, Validation & Governance -

• Experience building explainable AI outputs suitable for enterprise and governance use

cases.

• Ability to combine rule-based validation with LLM reasoning.

• Understanding of auditability, lineage, and reproducibility in AI-driven systems.

Good-to-Have:

• Experience with Data Vault 2.0 concepts (Hubs, Links, Satellites).

• Knowledge of Knowledge Graphs for entity and relationship grounding.

• Exposure to telecom or large-scale enterprise data domains.

• Integrate with:

1) Metadata systems (catalogs, schemas, lineage)

2) Transformation engines (dbt, SQL, Spark)

3) Quality frameworks (Great Expectations or similar)

• Collaborate on RAG + Knowledge Graph architectures for grounding LLM outputs.

• Optimize latency, cost, and reliability of LLM pipelines in production.