Senior AI Engineer: Agentic AI & LLM Systems

Convo; Islamabad; On-site; PKR 250,000 - 420,000; Full time

Job summary

Convo is seeking a Senior AI Engineer to design and build production-grade Agentic AI and LLM-powered systems. The ideal candidate is a strong Python engineer with hands-on experience in RAG, multi-agent architectures, knowledge graphs, and modern agent protocols.

You will design and implement scalable AI backends, integrate Neo4j and vector databases, and work with MA frameworks and APIs. This role offers the chance to impact enterprise workflows and collaborate with data, technology, and

Qualifications

• 6+ years of software engineering experience with significant AI/LLM work.
• Proven experience delivering production-grade AI or agentic systems.

Responsibilities

• Design and develop Agentic AI and multi-agent systems using Python.
• Build RAG, GraphRAG, knowledge graphs, and ontology-driven AI solutions.
• Develop agent workflows, routing, orchestration, and memory.
• Build high-performance AI backends and APIs with FastAPI.
• Integrate Neo4j, relational databases, vector databases with knowledge graphs.
• Work with MCP servers/clients and A2A protocols.
• Collaborate with MA Frameworks and other agent orchestration frameworks.
• Create tool-using agents interacting with APIs and enterprise systems.
• Design scalable, observable, production-ready AI architectures.

Skills

Python

Backend engineering

LLMs

RAG

Agentic AI

Multi-agent systems

FastAPI

Knowledge graphs

Neo4j

API development

System design

Testing

Scalability

Job description

Convo is seeking a Senior AI Engineer to design and build production-grade Agentic AI and LLM-powered systems. The ideal candidate is a strong Python engineer with hands-on experience in RAG, multi-agent architectures, knowledge graphs, and modern agent protocols.

You will design and implement scalable AI backends, integrate Neo4j and vector databases, and work with MA frameworks and APIs. This role offers the chance to impact enterprise workflows and collaborate with data, technology, and