Job Summary

The Agentic Orchestration Architect bridges established process-automation platforms and the new world of AI agents. The role designs business-process automation, workflow, RPA and human-in-the-loop architectures, and builds AI-agent, agentic workflow, MCP, LLM decisioning and autonomous orchestration solutions on top of them. It brings deep hands-on expertise in at least one enterprise automation platform (Camunda, Pega, Appian, Power Automate, UiPath, or Automation Anywhere), with working knowledge across the wider landscape.

As a CoE role, it defines reusable frameworks, governance, accelerators and best practices, evaluates emerging technology through PoCs, creates architecture and thought-leadership assets, and mentors teams. It also supports Solutions and Advisory teams with expert input to customer roadmaps, workshops and proposals.

Design business-process automation, workflow, RPA and human-in-the-loop architectures; build AI-agent, agentic workflow, MCP, LLM decisioning and autonomous orchestration solutions; deep hands-on expertise in at least one of Camunda, Pega, Appian, Power Automate, UiPath or Automation Anywhere, with working knowledge across the broader platform landscape; define reusable frameworks, governance, accelerators and best practices; evaluate emerging technology and build PoCs; shape customer roadmaps; create architecture and thought-leadership assets; mentor teams and support workshops and proposals. Mandatory across all roles: practical AI-native engineering; strong modern application development; cloud-native, API, microservices and DevSecOps grounding; hands-on experience with at least one agent harness such as Claude Code, GitHub Copilot, Codex, Kiro or Cursor; enterprise security, governance, evaluation, observability and cost awareness.

Key Responsibilities

A. Process Automation & Workflow Architecture

• Design business-process automation, workflow, and RPA architectures: process orchestration, case management, rules and decisioning, task automation, and integration.
• Decide the right mix of deterministic workflow, RPA, APIs, and AI agents for each process, based on variability, risk, volume, and cost.
• Design human-in-the-loop architectures: task routing, approvals, exception handling, escalation, SLAs, and audit trails.

B. Agentic & AI-Driven Solutions

• Build AI-agent, agentic workflow, and autonomous orchestration solutions, including agents embedded in or coordinated by process platforms.
• Design LLM decisioning patterns: classification, extraction, routing, recommendation, and judgement tasks with confidence thresholds, fallbacks, and explainability.
• Design and integrate MCP servers and tools so agents and workflows can safely invoke enterprise systems and capabilities.
• Define autonomy levels and guardrails per process step, with controls for when work must be handed to a human.

C. Platform Expertise & Integration

• Provide deep, hands-on architecture and build leadership on at least one of Camunda, Pega, Appian, Power Automate, UiPath, or Automation Anywhere.
• Maintain working knowledge across the broader landscape (BPM/BPMN engines, low-code platforms, RPA suites, iPaaS, document processing, process mining, decision engines) and guide platform selection and coexistence.
• Architect API, event-driven, and microservices integration with systems of record and legacy applications, including UI-level automation where no API exists.

D. Reusable Frameworks, Governance & Accelerators

• Define reusable frameworks, reference architectures, and accelerators: process templates, agent patterns, connector and MCP templates, exception-handling patterns, and evaluation harnesses.
• Establish governance and best practices: automation and agent lifecycle management, center-of-excellence operating models, change control, bot and agent identity, access, and audit.
• Define evaluation, observability, security, and cost controls for automations and agents (quality metrics, tracing, SLOs, least privilege, token and licence cost management).

E. Innovation & Proofs of Concept

• Evaluate emerging technology (agent frameworks, model capabilities, platform AI features, interoperability protocols) and recommend where it fits.
• Build PoCs that prove feasibility and value; convert successful PoCs into reusable assets and delivery-ready patterns.
• Use AI-native engineering and agent harnesses to build and iterate quickly while maintaining production quality.

F. Field Enablement & Engagement Support

• Provide expert architecture input to customer automation and agentic roadmaps led by Solutions and Advisory teams (target architecture, platform guidance, patterns to reuse).
• Support workshops, discovery sessions, and demos as a technical subject-matter expert.
• Contribute architecture content, estimation inputs, and risk framing to proposals and pursuits.
• Turn field feedback and engagement learnings back into CoE frameworks and accelerators.

G. Thought Leadership & Mentoring

• Create architecture and thought-leadership assets: whitepapers, point-of-view documents, reference diagrams, blogs, conference and client-briefing material.
• Mentor engineers and architects; build capability across platform and agentic skills; contribute to training and certification pathways.

Skill Requirements

• Bachelor's or Master's degree in Computer Science, Engineering, Information Systems, or an equivalent field, or equivalent practical experience.
• 12+ years in software, process automation, or enterprise architecture, with at least 6 years on one or more of the named automation platforms and at least 2 years building AI-agent or LLM-based solutions.
• Track record of delivering production automation or workflow solutions at enterprise scale, and of supporting field teams in customer workshops or proposals.

Technical Skills (Must Have)

1. Process and workflow architecture: BPMN/DMN, orchestration vs. choreography, case management, rules and decisioning, process design patterns.
2. Enterprise automation platform depth: deep hands-on expertise in at least one of Camunda, Pega, Appian, Power Automate, UiPath, or Automation Anywhere, including design, build, deployment, and governance.
3. Broader platform landscape: working knowledge of other BPM, low-code, RPA, iPaaS, document processing, and process-mining platforms, and when to use each.
4. RPA and integration: attended and unattended automation, UI and API automation, connectors, legacy integration, resilience and exception handling.
5. Human-in-the-loop design: task management, approvals, escalation, SLA management, audit trails.
6. AI agents and agentic workflows: agent architecture, orchestration, tool use, planning, workflow vs. agent trade-offs, autonomous orchestration with controls.
7. LLM decisioning: prompt and context engineering, structured outputs, confidence handling, fallbacks, explainability.
8. MCP and tool integration: designing and consuming MCP servers; secure exposure of enterprise capabilities to agents.
9. AI-native engineering: practical use of AI across the development lifecycle; hands-on with at least one agent harness (Claude Code, GitHub Copilot, Codex, Kiro, or Cursor).
10. Modern application development: strong proficiency in at least one language such as Python, TypeScript, or Java; clean architecture, testing, code review.
11. Cloud-native, API & microservices: AWS/Azure/GCP, containers, REST/event-driven integration, service design.
12. DevSecOps: CI/CD for automation assets, infrastructure as code, secrets and credential management, secure SDLC.
13. Enterprise readiness: security, governance, evaluation, observability, and cost awareness, including token, licence, and run-cost modeling for automations and agents.

Behavioral & Professional Skills

• Translates business process pain points into reusable architecture patterns and value cases.
• Clear communication and facilitation with business leaders, process owners, and engineers; strong written and visual architecture communication.
• Comfortable bridging traditional automation and emerging AI approaches without favouring either by default.
• Mentoring, collaboration, and a strong sense of ownership across Advisory, Solutions, and CoE teams.

Other Requirements

Role-specific

• Process mining and task mining (Celonis, UiPath Process Mining, Microsoft Process Mining).
• Intelligent document processing and OCR/extraction platforms.
• Decision management and rules engines (DMN, Pega Decisioning, Drools, and similar).
• Agent frameworks and runtimes (LangGraph, Claude Agent SDK, OpenAI Agents SDK, Semantic Kernel, AutoGen, CrewAI) and agent interoperability protocols (A2A).
• Platform-native AI features (e.g., Pega GenAI, UiPath agentic capabilities, Appian AI, Copilot Studio, Camunda AI).
• Durable execution engines (Temporal, Step Functions, Durable Functions).
• Automation CoE operating models, FinOps for automation, and business case/ROI modeling.
• Domain exposure in BFSI, insurance, healthcare, telecom, manufacturing, or public sector.

Common competencies (mandatory across all roles)

• Practical AI-native engineering and strong modern application development.
• Cloud-native, API, microservices, and DevSecOps grounding.
• Hands-on experience with at least one agent harness (Claude Code, GitHub Copilot, Codex, Kiro, Cursor).
• Enterprise security, governance, evaluation, observability, and cost awareness.

Certifications (preferred, not mandatory)

• Platform certifications: Camunda, Pega (CSA/CSSA/LSA), Appian (Certified Lead Developer), Microsoft Power Platform (PL-400/PL-600), UiPath (Advanced Developer/Solution Architect), Automation Anywhere (Advanced RPA Professional).
• Cloud solution architect certifications (AWS, Azure, or GCP).
• TOGAF or equivalent architecture certification.