Senior Neuro-Symbolic AI Architect

Indian AI Research Organization (IAIRO); Bandon; On-site; EUR 120,000 - 180,000; Full time

Job summary

Indian AI Research Organization (IAIRO) is seeking a Senior Software Engineer to architect and implement hybrid AI systems that combine LLMs with structured knowledge. You will train, fine-tune, and optimize large-scale models, integrating them with Knowledge Graphs and ontologies to deliver grounded, production-ready solutions.

You will lead a team of engineers and researchers, driving end-to-end development from model training to deployment on high-performance compute resources.

Qualifications

• MS or PhD in Computer Science, AI, or related field or equivalent industry experience.
• 5+ years software engineering with Python proficiency.
• Extensive LLM training and RAG architectures experience.
• Strong knowledge of Knowledge Graphs, SPARQL/Cypher and RDF/OWL.

Responsibilities

• Design and implement neuro-symbolic integration architectures linking LLMs with knowledge graphs.
• Lead LLM training, fine-tuning, and optimization on domain-specific data.
• Build scalable knowledge representations, ontologies and triple stores as truth layer.
• Develop high-performance backends for neuro-symbolic AI with low-latency inference.
• Translate cutting-edge research into robust production-ready code.
• Mentor engineers and researchers to foster rigorous engineering and AI development.

Skills

Python

LLM training

Knowledge graphs

PyTorch

TensorFlow

Distributed training

CUDA

Education

MS/PhD in CS/AI

Tools

SPARQL

Cypher

RDF/OWL

DeepSpeed

Megatron-LM

CUDA

Job description

Indian AI Research Organization (IAIRO) is seeking a Senior Software Engineer to architect and implement hybrid AI systems that combine LLMs with structured knowledge. You will train, fine-tune, and optimize large-scale models, integrating them with Knowledge Graphs and ontologies to deliver grounded, production-ready solutions.

You will lead a team of engineers and researchers, driving end-to-end development from model training to deployment on high-performance compute resources.