Head of Neuro-Symbolic AI

Sapience AI Corporation•Seattle, WA

1d•$216,000 - $240,000

About The Position

Sapience AI is seeking a Head of Neuro-Symbolic AI to lead the development and direction of its core reasoning architecture, COGENT. This role is crucial for combining symbolic knowledge, structure, and rules with neural and language-model capabilities to enable the platform to reason over community expertise with both flexibility and rigor. The Head of Neuro-Symbolic AI will own the architectural direction, lead a specialized team, and partner across research, applied AI, and knowledge graph engineering to create a trustworthy, explainable, and traceable reasoning system at the heart of the MINERVA platform. This position is not a pure research role, a general engineering leadership role, a knowledge-graph-only role, or a benchmark-chasing role, but is accountable for reasoning that works in production and is measured by outcomes such as trustworthy reasoning, a working architecture, explainability, a strong team, production-ready advances, and honest measurement.

Requirements

Deep expertise in neuro-symbolic AI, knowledge representation, reasoning, or a closely related field.
Experience combining symbolic methods with neural or language-model systems.
A track record of turning advanced AI research into working systems.
Strong grounding in knowledge graphs, retrieval, and modern LLMs.
Experience leading researchers or engineers, or clear readiness to lead.
Rigor about evaluation, groundedness, and honest handling of uncertainty.
The ability to translate hard ideas for a broad technical audience.
Knowledge graph and symbolic reasoning systems, ontologies, and rule engines.
Modern ML frameworks (PyTorch or equivalent) and LLM tooling.
Retrieval, embeddings, and vector systems.
Graph databases (for example Neo4j-class systems).
Evaluation and experiment-tracking tooling.
Python as the primary language.
Ownership of the COGENT architecture and its integration with the KO graph and MINERVA.
Prior senior work in neuro-symbolic AI, knowledge-based systems, or applied reasoning research.
Experience shipping reasoning or knowledge systems in production.
A background bridging research and engineering.

Nice To Haves

A graduate degree or equivalent depth in AI, machine learning, computational logic, or a related area.
Publications, patents, or shipped systems in neuro-symbolic or knowledge-based AI.
Experience with explainable AI and provenance.
Experience productionizing reasoning systems at scale.
Domain understanding of professional communities or knowledge-intensive organizations.
Experience leading a team or a significant technical area is a plus.

Responsibilities

Own the direction of the COGENT architecture and how it fuses symbolic reasoning with neural and language-model capability.
Set the approach to grounding, structure, and rules that keeps reasoning trustworthy.
Decide what belongs in the symbolic layer, the neural layer, and the seam between them.
Own how COGENT reasons over the KO graph to turn a community’s expertise into answers members can act on.
Partner with knowledge graph engineering on the structures reasoning depends on.
Advance retrieval, inference, and symbolic reasoning together, not in isolation.
Make reasoning explainable and traceable, so members can see why the platform says what it says.
Build the guardrails and provenance that let communities trust and verify answers.
Reduce confident errors, and handle uncertainty honestly.
Set the research agenda for neuro-symbolic methods at Sapience AI.
Decide which advances are ready for production and which need more work.
Keep the approach current with the field without chasing every trend.
Build and lead a specialized team of researchers and engineers.
Set a high bar for rigor and for shipping, and hold both.
Grow the people who will define this area for the company.
Partner with platform and applied AI to run neuro-symbolic reasoning reliably and affordably in production.
Balance ambition against the constraints of latency, cost, and scale.
Turn architectural advances into dependable behavior in MINERVA.
Own how the quality of reasoning is measured, including accuracy, groundedness, and trust.
Build evaluation that reflects what communities actually need, not just benchmark scores.
Use evidence to steer the architecture’s direction.