Agentic AI Architect — Orchestrating Multi-Agent Systems

iSystems Ltd; Lahore; On-site; PKR 4,000,000 - 8,000,000; Full time

Job summary

iSystems Ltd. seeks a senior AI/Software Architect to design and govern agentic, multi-agent workflows. You will lead architecture decisions, memory architectures, and tool-use governance for reliable autonomous systems. The role requires mentoring engineers and presenting tradeoffs to clients.

You will collaborate with Security, Governance, and Platform Engineering to ensure safe, scalable agent-based solutions across client engagements.

Qualifications

• 10-15+ yrs software/AI architecture experience with 2+ yrs in agentic/multi-agent systems.
• Deep knowledge of agent orchestration frameworks and failure modes (loops, deadlocks, drift).
• Experience designing memory architectures for long-running agents.
• Writes clear ADRs and architecture documentation for engineers and executives.
• Experience mentoring/leading a technical team and owning architecture decisions.

Responsibilities

• Design multi-agent architectures and handoff logic (LangGraph, AutoGen, CrewAI).
• Own agentic workflow design: memory/state management and knowledge graphs.
• Define guardrails: tool-use permissioning, kill switches, escalation paths.
• Architect evaluation framework for agentic systems with QA.
• Lead technical design reviews before build.
• Design fallback and human-in-the-loop patterns for high-stakes decisions.
• Mentor GenAI Engineers on architecture patterns.
• Present tradeoffs to client stakeholders.
• Own architecture governance across the practice.
• Facilitate cross-team design reviews with Security and Platform Engineering

Skills

Agentic systems

Multi-agent architectures

Distributed systems

Memory architectures

ADR documentation

Team mentoring

Guardrails tooling

Architectural reviews

Client presentations

Tools

Azure AI Foundry Agent Service

AWS Bedrock Agents

Vertex AI Agent Builder

LangGraph

AutoGen

CrewAI

Job description

iSystems Ltd. seeks a senior AI/Software Architect to design and govern agentic, multi-agent workflows. You will lead architecture decisions, memory architectures, and tool-use governance for reliable autonomous systems. The role requires mentoring engineers and presenting tradeoffs to clients.

You will collaborate with Security, Governance, and Platform Engineering to ensure safe, scalable agent-based solutions across client engagements.