Data Scientist

About Demandbase:

Demandbase is the Smarter GTM™ company for B2B brands. We help marketing and sales teams overcome the disruptive data and technology fragmentation that inhibits insight and forces them to spam their prospects. We do this by injecting Account Intelligence into every step of the buyer journey, wherever our clients interact with customers, and by helping them orchestrate every action across systems and channels - through advertising, account-based experience, and sales motions. The result? You spot opportunities earlier, engage with them more intelligently, and close deals faster.

As a company, we’re as committed to growing careers as we are to building world-class technology. We invest heavily in people, our culture, and the community around us. We have offices in the San Francisco Bay Area, New York, Seattle, and teams in the UK and India, and allow employees to work remotely. We have also been continuously recognized as one of the best places to work in the San Francisco Bay Area.

We're committed to attracting, developing, retaining, and promoting a diverse workforce. By ensuring that every Demandbase employee is able to bring a diversity of talents to work, we're increasingly capable of living out our mission to transform how B2B goes to market. We encourage people from historically underrepresented backgrounds and all walks of life to apply. Come grow with us at Demandbase!

About the Role

We are looking for a Senior Data Scientist to help architect and build next-generation Agentic AI systems at Demandbase. This role focuses on multi-agent orchestration, LLM-powered reasoning systems, evaluation frameworks, guardrails, and scalable GenAI architectures.

You will work at the intersection of advanced data science, generative AI research, and production-grade ML systems, shaping how intelligent agents operate reliably, safely, and effectively in enterprise environments.

This is not a platform infrastructure role — it is a deep AI systems engineering role centered around agent architecture, model evaluation, reasoning systems, and applied ML innovation.

Key Responsibilities

Agentic AI & Multi-Agent Architecture

• Design and implement multi-agent systems for complex enterprise workflows.
• Build agent orchestration frameworks (planner-executor, tool-using agents, retrieval-augmented agents, self-reflective agents).
• Develop architectures for reasoning loops, memory systems, tool integration, and contextual grounding.
• Design guardrails for hallucination mitigation, tool misuse prevention, and safe execution.
• Implement feedback-driven refinement loops and self-correction strategies.

GenAI Systems, Evals & Guardrails

• Design and operationalize LLM evaluation frameworks (automated evals, LLM-as-judge, human-in-the-loop, adversarial testing).
• Build robust prompt engineering and prompt versioning strategies.
• Develop safety guardrails including content filtering, policy enforcement, and bias monitoring.
• Implement quality metrics for:
• Factual accuracy
• Groundedness
• Latency and cost efficiency
• Agent reliability
• Create structured evaluation pipelines to continuously improve agent performance.

Advanced Data Science & NLP

• Apply advanced NLP techniques (transformers, embeddings, fine-tuning, RAG pipelines).
• Work deeply with unstructured and semi-structured data.
• Develop model experimentation frameworks for prompt optimization, fine-tuning, and retrieval strategies.
• Optimize data pipelines using Python, Pandas, Spark, and vector databases.
• Collaborate with data scientists to convert research prototypes into scalable AI systems.

AI System Design & Architecture

• Architect modular, extensible AI systems for long-term maintainability.
• Design retrieval-augmented generation (RAG) systems with advanced chunking, embedding strategies, and re-ranking.
• Build memory architectures (short-term, long-term, vector-based).
• Optimize inference pipelines for performance, cost, and reliability.
• Define reusable patterns for enterprise-grade AI systems.

Operational Excellence for AI Systems

• Implement evaluation-driven CI/CD for GenAI systems.
• Establish monitoring for:
• Model drift
• Agent failure modes
• Tool misuse
• Hallucination frequency
• Maintain reproducibility through experiment tracking and versioning.
• Ensure ethical AI practices and compliance with enterprise standards.

Technical Leadership & Mentorship

• Define best practices for agentic AI architecture and LLM system design.
• Mentor engineers and data scientists in advanced GenAI methodologies.
• Drive internal thought leadership in Agentic AI and applied GenAI research.
• Contribute to building a high-performing AI engineering culture.

Basic Qualifications

• 8+ years of experience in Machine Learning, AI Engineering, or Applied Data Science.
• Strong expertise in Generative AI systems and LLM architectures.
• Experience designing multi-agent or tool-using AI systems.
• Deep proficiency in Python and ML ecosystems (NumPy, Pandas, PyTorch/TensorFlow).
• Hands-on experience with:
• RAG systems
• Prompt optimization
• Evaluation frameworks
• Embedding models & vector databases
• Strong understanding of transformers, embeddings, and fine-tuning methods.
• Experience building production-grade AI systems from research to deployment.
• Strong system design and architectural problem-solving skills.

Preferred Qualifications

• MS or PhD in Computer Science, AI, ML, or related fields.
• Experience with LLM orchestration frameworks (LangChain, LlamaIndex, custom agent frameworks).
• Experience with automated eval frameworks and benchmarking strategies.
• Familiarity with reinforcement learning, RLHF, or agent self-improvement loops.
• Experience with vector databases and retrieval systems.
• Published research or open-source contributions in AI/ML.
• Strong understanding of AI safety and responsible AI practices.

What You’ll Gain

• Opportunity to architect enterprise-scale Agentic AI systems.
• Direct impact on next-generation AI product capabilities.
• Exposure to cutting-edge developments in LLMs, multi-agent systems, and AI safety.
• A culture that values deep technical thinking, experimentation, and innovation.
• The chance to help define Demandbase’s AI future.