Artificial Intelligence Engineer

At Central Group Digital & AI Team, we build reusable AI/ML, GenAI, and agentic

products & solutions that companies across the group adopt and run in production.

You will build the intelligence at the core of them agentic systems and

knowledge/context graphs frameworks and take them all the way to production

with clean, reliable, production-grade code.

What you will do.

• Build reusable AI/GenAI products centred on agentic systems and

knowledge/context graphs

• Design multi-agent workflows: orchestration, tool use, memory, planning,

guardrails, and human-in-the-loop

• Build knowledge/context graphs that ground agents and LLMs GraphRAG,

ontologies, and entity/relationship extraction & resolution

• Ship production-grade services, well-tested, observable, evaluated via

Docker, CI/CD, and cloud, designed for reuse through configuration rather

than rewrite

• Partner with group companies and AI and data teams to take solutions from

prototype to dependable production

Tech

Python · Agentic AI orchestration, tool calling, memory, guardrails

(LangGraph/LangChain, CrewAI, AutoGen) · Knowledge/context graphs -

Neo4j/Cypher, GraphRAG, ontology & knowledge modelling, entity/relationship

extraction · RAG + vector DBs (pgvector/Pinecone/Weaviate/Qdrant) · model APIs +

Hugging Face, PyTorch, scikit-learn · FastAPI, REST, microservices · Docker,

Kubernetes, CI/CD, cloud (AWS/GCP/Azure), monitoring/observability · PostgreSQL,

Fluent with AI tools for coding and data analysis such as Claude, Copilot, and

Cursor.

You bring

• 5–7 years building and shipping ML/AI systems to production with production-grade, well-tested code
• Strong Python and hands-on agentic AI orchestration, tool calling, memory, guardrails, and agent frameworks
• Real experience with knowledge/context graphs, graph databases such as Neo4j, GraphRAG, ontology/knowledge modelling, entity and relationship extraction
• Solid RAG, vector databases, evaluation, and LLMOps practice. Comfort owning things in production Docker, CI/CD, cloud, monitoring and not just prototypes

Experience building at a product start-up is a strong plus: shipping agentic and

AI systems from zero to production, owning them end to end, and moving fast with a

small team is exactly the mindset we want.

Bonus: fine-tuning / model optimisation · semantic web (RDF/SPARQL), exposure to

industrial AI, IoT, digital twins, or enterprise workflow automation (a plus, not

required)

Education: Bachelor's/Master's in Computer Science, Engineering, Data Science

from reputed Institute