We are looking for a Data Scientist with hands-on Agentic AI experience to join our team and help build intelligent, data-driven solutions on Google Clouds Vertex AI platform.

You will combine rigorous data science practices modelling, experimentation, and analysis with the ability to design and integrate AI agents into real-world workflows. This is a unique role at the intersection of classical data science and next-generation AI.

Key Responsibilities :

- Develop and deploy ML models on Vertex AI for use cases such as predictive analytics, anomaly detection, classification, and forecasting.

- Architect end-to-end agentic AI solutions on Vertex AI, including agent orchestration, tool use, memory, and multi-agent coordination.

- Integrate the agentic setup with existing infrastructure: Cloud Functions, Cloud Run, and Firestore.

- Define agent workflows planning loops, reasoning strategies, tool calling, and human-in-the-loop escalation paths.

- Build and maintain RAG pipelines including data ingestion, chunking, embedding, and retrieval to ground agents in enterprise data.

- Run experiments and evaluations to benchmark model and agent performance, iterating based on rigorous metrics.

- Translate business problems into well-defined data science and AI problems with measurable outcomes.

- Document methodologies, model cards, and agent design decisions for transparency and reproducibility.

Required Skills & Experience :

- 7+ years of data science experience, including ML model development, evaluation, and production deployment.

- Hands-on experience with agentic AI concepts tool/function calling, RAG, LLM orchestration, or multi-agent workflows.

- Proficiency in Python (pandas, scikit-learn, PyTorch or TensorFlow) and data science tooling.

- Experience with Vertex AI including Vertex AI Workbench, Pipelines, Model Registry, and/or Vector Search.

- Strong understanding of statistical modelling, feature engineering, and experiment design.

- Familiarity with LLM prompting strategies, structured outputs, and evaluation frameworks (e.g., RAGAS, LangSmith).

- Experience with cloud data infrastructure BigQuery, Cloud Storage, or similar.

- Good communication skills able to explain model behaviour and agent decisions to non-technical stakeholders.

Nice to Have :

- Experience with LangChain, LangGraph, CrewAI, or Googles Agent Development Kit (ADK).

- Familiarity with Firestore as an agent memory or state store.

- Exposure to industrial or operational data (IoT, time-series, supply chain, plant operations).

- Experience with fine-tuning or RLHF on Gemini or open-source LLMs.

- Google Cloud certification: Professional ML Engineer or Professional Data Engineer.