Head of Data Science

Our client is building an AI-native data infrastructure platform from the ground up. This is a true 0-1 build - no legacy stack, no legacy thinking.

We are looking for a Head of Data Science to be the founding data science leader. This is a player-coach role for a builder who combines deep technical depth with extreme ownership and a bias for shipping.

You will own the data science charter end-to-end: from defining the initial architecture and use-cases, to building the first team, to deploying AI-native capabilities that become core to the platform itself. You will work directly with Founders / C-suite and Engineering leadership.

If you want to architect how modern companies build on data with AI at the core, this is that role.

A. ROLE MANDATE

1. 0-1 Platform Leadership: Define and own the data science vision, strategy, and roadmap for an AI-native data infra platform.
2. Build & Scale the Function: Be the founding leader - hire, mentor, and scale a high-caliber team of Data Scientists and ML Engineers. Set the culture, bar, and operating principles from day one.
3. Ship AI into the Core Product: Not AI as a feature - AI as the foundation. Design intelligent systems that power automation, discovery, governance, quality, and decisioning within the data stack itself.

B. ROLES & RESPONSIBILITIES

1. Strategy & 0-1 Execution

• Translate ambiguous 0-1 product vision into a concrete data science roadmap with clear milestones and business impact.
• Own projects from conception -> prototype -> production -> iteration in a fast-moving environment.
• Partner with Product and Engineering to make critical build vs. buy, architecture, and modeling decisions for the platform.

2. AI-Native Solution Development

• Design, build, and deploy production-grade ML / GenAI systems that are native to the data infrastructure layer - e.g. intelligent data discovery, auto-optimization, anomaly detection, semantic layer, LLM-powered data agents, and self-healing pipelines.
• Lead full lifecycle development: data exploration, feature engineering, model training/evaluation, deployment, monitoring for drift/performance, and continuous retraining.
• Establish MLOps / LLMOps best practices from scratch: model registry, versioning, evaluation frameworks, observability, and governance.

3. Architecture & Infrastructure Partnership

• Co-architect the underlying data platform with Data Engineering - ensuring scalability, reliability, and cost-efficiency for training and inference at scale.
• Evaluate and implement modern stack components: vector databases, feature stores, orchestration, LLMs / SLMs, RAG frameworks, knowledge graphs.
• Define and own success metrics for all data science initiatives.

4. Leadership & Evangelism

• Act as a strategic thought partner to leadership, translating complex technical concepts into clear business decisions.
• Champion excellent data science practices and a culture of experimentation, rigor, and documentation.
• Stay at the forefront of AI/ML research and rapidly assess practical application to the platform.

C. WHO WE ARE LOOKING FOR

This is for a builder, not a manager of a large existing team.

Experience:

• 8+ years in Data Science / Applied ML, with at least 3+ years leading teams or as a senior Tech Lead in a 0-1 or high-growth environment.
• Proven track record of taking ML / GenAI products from whiteboard to scaled production with measurable impact. You have built something from scratch.
• Experience building or scaling a data platform, infra platform, or AI platform product is a massive plus. B2B SaaS / Data Infra background preferred.

Technical Depth:

• Expert-level in Python, SQL, and core ML libraries. Strong in at least one deep learning framework (PyTorch, TensorFlow).
• Deep expertise in at least TWO of: Recommender Systems, NLP / LLMs / RAG / Agents, Knowledge Graphs, Time-series / Anomaly Detection, Large-scale Optimization.
• Strong fundamentals: statistics, experimental design, evaluation, feature engineering, model selection.
• Hands-on with modern data stack: Spark, dbt, Airflow/Dagster, Snowflake/BigQuery/Databricks, vector DBs (Pinecone, Weaviate, pgvector), MLOps (MLflow, Weights & Biases).
• Comfortable with ambiguity and complex, high-dimensional data.

Mindset:

• Founder mentality: Extreme ownership, high agency, hands-on when needed.
• Product-minded scientist - obsessed with delivering value, not just model accuracy.
• Excellent communicator who can influence both deeply technical and non-technical stakeholders.

Education:
Master's / PhD in Computer Science, Machine Learning, Statistics, Mathematics or related field preferred, but exceptional track record and real-world shipped products trump degrees.