AI Workflow Engineering Architect

AppliedAI / Opus
Abu Dhabi
Workplace: OnsiteFull timeFunction: Software EngineeringSkills: ["Mentorship","Technical leadership","Standard-setting","Architecture reviews","Production troubleshooting"]

Own the technical architecture of production workflows within Opus, focusing on ML-intensive, latency- and cost-aware execution. Lead decomposition of business processes into workflow graphs, define model routing and evaluation standards, and drive systems-level performance analysis (caching, batching, concurrency, inference). Establish reliability and execution semantics, instrument production feedback loops, and mentor Workflow Engineers to institutionalize workflow engineering best practices.

Loading

Loading job details...

Preparing the role view and application actions.

FursaFursa
AppliedAI / Opus
AppliedAI / Opus
2 hours ago

AI Workflow Engineering Architect

✓ Verified Job

Canonical indexed version, validated from employer's careers page.

Source: Company careers pageValidated by: Fursa AI
Last checked: 2 hours agoStatus: Live

Job Summary

Own the technical architecture of production workflows within Opus, focusing on ML-intensive, latency- and cost-aware execution. Lead decomposition of business processes into workflow graphs, define model routing and evaluation standards, and drive systems-level performance analysis (caching, batching, concurrency, inference). Establish reliability and execution semantics, instrument production feedback loops, and mentor Workflow Engineers to institutionalize workflow engineering best practices.
Location: Abu Dhabi
Workplace: Onsite
Employment Type: Full time
Job Function: Software Engineering
Seniority: Mid level

Key Responsibilities

  • •Lead decomposition of business processes into efficient Opus workflow graphs, defining execution boundaries, dependencies, state, parallelism, and failure paths.
  • •Establish technical methods for model selection and composition within Opus, including routing, cascades, conditional strategies, and human review.
  • •Define engineering standards for workflow evaluation, including representative test sets, business-weighted loss functions, benchmarks, ablation methods, and distribution-shift coverage.
  • •Perform technical analysis of inference and workflow execution performance (cost, context construction, retrieval, serialization, network calls, concurrency, orchestration overhead) and drive improvements.
  • •Set reliable workflow execution semantics (typed interfaces, state transitions, idempotency, checkpointing, retries/timeouts, backpressure, and recovery) and lead production measurement and continuous optimization.

Key Requirements

  • •Senior-level applied ML and software engineering experience designing and operating production LLM/agentic systems (not research-only or prototypes).
  • •Strong grounding in ML evaluation methodology and systems performance across latency, cost, and concurrency.
  • •Distributed systems fundamentals including reliability, failure handling, and state management in production pipelines.
  • •Experience mentoring engineers and setting technical standards, not only executing work solo.
Experience:MLLLMsAgentic systemsDistributed systemsProduction engineering
Skills:MentorshipTechnical leadershipStandard-settingArchitecture reviewsProduction troubleshooting
Tech Stack:LLMsAgentic systemsAnthropicOpenAIGeminiRetrievalCascadesConditional routingCheckpointingProfilingCachingBatchingConcurrencyOrchestrationQuantizationTyped interfaces

Company Brief

AppliedAI / Opus
AppliedAI builds Opus, an agentic AI workflow platform for regulated enterprises. It helps organizations discover, build, run, and optimize governed business processes with human oversight, auditability, and compliance controls, especially in banking, healthcare, insurance, and other regulated sectors.
Industry: Enterprise Software
Company Size: Large (251 to 1,000 employees)
Growth: Scaleup
Funding: Series B
Headquarters: Abu Dhabi, United Arab Emirates
Founded: 2021
WebsiteLinkedIn