R&D Engineer

Pioneer the future of decision intelligence as an R&D Engineer at Rainbird, blending the precision of symbolic AI with the power of large language models. You’ll innovate at the cutting edge of neurosymbolic integration, architecting advanced AI systems that deliver explainability, determinism, and accuracy. Collaborate with leading experts to shape technology that transforms complex decision-making.

Contract Type: Permanent, Full Time
Location: Hybrid (Norwich / London) or Remote (UK)
Package: Competitive, based on experience

About Rainbird

Rainbird Technologies is an innovative artificial intelligence scale-up based in Norwich. We empower organisations to automate complex decision-making using our award-winning low-code SaaS platform.

We are looking for an R&D Engineer to help advance our neurosymbolic AI engine, integrating the natural language capabilities of large language models (LLMs) with the logical reasoning power of symbolic AI. This role is crucial in developing the next generation of decision intelligence for high stakes applications, where explainability, determinism, and precision are key.

Role Specification

As an R&D Engineer at Rainbird, you’ll be at the forefront of integrating probabilistic and symbolic AI. Your work will bridge the gap between the probabilistic, pattern-matching capabilities of LLMs and the logical precision of symbolic reasoning systems.

You will collaborate directly with our core engineering team, Head of Engineering, and CTO to architect novel approaches that leverage the complementary strengths of these technologies. This involves designing systems where LLMs can effectively communicate with our knowledge graph infrastructure, translating natural language into structured symbolic representations and vice versa.

A significant part of your role will focus on extending our platform’s capabilities by developing algorithms and solutions that manage the interaction between different AI paradigms. This includes creating mechanisms for knowledge extraction and transfer between LLMs and symbolic systems. You’ll design and implement neurosymbolic architectures that preserve the interpretability advantages of symbolic AI while incorporating the flexibility of neural networks.

Beyond the neurosymbolic integration work, you’ll contribute to broader innovation around our core platform. You’ll explore emerging technologies and methodologies that could enhance Rainbird’s capabilities in areas such as automated knowledge acquisition, reasoning transparency, and computational efficiency. This requires staying current with academic research and industry developments to identify opportunities for platform evolution.

The role demands expertise in LLM optimization techniques including fine-tuning on domain-specific data, crafting robust prompting strategies, and implementing retrieval-augmented generation style architectures. You’ll apply these techniques to enhance decision accuracy while maintaining deterministic behavior where required.

Your work will often focus on proof-of-concept implementations and technical prototypes that demonstrate feasibility and value. These innovations will feed into our core product roadmap, where our product engineering teams will transform your research into production-ready features. You’ll provide technical guidance during this transition to ensure the essence of your innovations is preserved.

Finally, you’ll establish rigorous evaluation frameworks to assess the reasoning capabilities of hybrid systems. This involves designing benchmark tests, measuring logical soundness, identifying edge cases, and creating metrics that help product teams understand the strengths and limitations of different approaches.

Requirements

We are seeking a candidate with a robust technical foundation and practical experience in neurosymbolic AI. The ideal candidate will possess:

• An advanced degree (Master’s or Ph.D.) in Computer Science, Artificial Intelligence, Machine Learning, or a related field, demonstrating a solid foundation in AI principles and methodologies.
• Proficiency in programming languages, particularly Python, and experience with a strongly typed language such as Go, enabling the development of robust and efficient codebases.
• A strong understanding of AI, machine learning, or computational reasoning, with hands-on experience in symbolic AI techniques, knowledge representation, and rule-based systems.
• Experience with large language models (LLMs) and associated tooling( for example OpenAI, Anthropic, Huggingface), and a solid grasp of natural language processing techniques to enhance machine understanding and interaction.
• Knowledge of vector databases, embeddings, and retrieval-augmented generation style architectures.
• A proven track record of conducting rigorous research and translating theoretical findings into practical, solutions that drive value.
• Exceptional analytical and problem-solving skills, coupled with a relentless passion for driving innovation within the AI landscape.

Preferred Experience

Candidates who stand out will also bring:

• Experience in creating integrated AI systems that blend symbolic and neural methodologies, pushing the boundaries of conventional AI applications.
• Exposure to graph-based reasoning and the construction and utilization of knowledge graphs, facilitating sophisticated data relationships and inferencing.
• A background in research and development-focused roles or active participation in academic-industry collaborations, showcasing a commitment to advancing the field through shared knowledge and innovation.

Why Join Us?

• Work at the cutting edge of AI innovation, pioneering neuro-symbolic intelligence.
• Collaborate with an ambitious and highly skilled team.
• Fully remote work with flexible arrangements.
• Opportunity to contribute to groundbreaking advancements in explainable AI.

Interested candidates should apply below and submit their CV and a brief covering letter outlining relevant experience. We look forward to hearing from you!

Rainbird is an equal opportunities employer.

#J-18808-Ljbffr