AI Workflow Engineering Architect

Position Overview

Workflow Architecture and Optimization

• Decompose business processes into efficient Opus workflow graphs with execution boundaries, dependencies, state, parallelism, and failure paths.
• Optimize workflow decisions across accuracy, latency, throughput, inference cost, and operational reliability.

Model Architecture and Routing

• Establish methods for selecting and composing models within Opus.
• Design heterogeneous execution strategies using deterministic computation, specialist models, retrieval, cascades, routing, early exits, and human review.
• Support model and routing decisions with measured error rates, calibrated confidence, workload characteristics, and inference economics.

Evaluation and Performance

• Define standards, test sets, loss functions, benchmarks, ablation methods, confidence intervals, and regression thresholds for workflow evaluation.
• Analyze inference, retrieval, serialization, network, concurrency, and orchestration costs.
• Improve caching, batching, parallel execution, model size, quantization, context length, and accelerator utilization.
• Characterize realistic concurrency using throughput, cost per successful execution, and p50, p95, and p99 latency.

Reliability and Production Measurement

• Establish reliable execution practices including typed interfaces, state transitions, idempotency, checkpointing, bounded retries, timeouts, backpressure, compensation, and recovery.
• Apply property-based testing, fault injection, deterministic replay, and execution-trace analysis where appropriate.
• Instrument workflow versions, model choices, traces, errors, latency, inference consumption, and business outcomes.
• Lead regression diagnosis and determine the appropriate corrective action across workflow, data, models, routing, context, software, or capacity.

Technical Leadership

• Develop Workflow Engineers into independent machine learning and computer science practitioners.
• Lead difficult workflow designs, architecture reviews, post-mortems, reference implementations, engineering standards, and reusable Opus patterns.
• Train the team in optimization, evaluation, inference engineering, and production machine learning.

Required Skills

• Senior-level applied ML and software engineering background with hands-on production LLM or agentic systems experience.
• Strong grounding in ML evaluation methodology and systems performance, including latency, cost, and concurrency.
• Distributed systems fundamentals covering reliability, failure handling, and state management in production pipelines.
• Experience mentoring engineers and setting technical standards.

Desirable Skills

• Experience with multi-model orchestration, routing, cascades, and provider fallbacks.
• Background in healthcare, finance, insurance, BPO, or another regulated industry.
• Experience building engineering standards or a technical discipline in an early-stage environment.
• Published writing, research, or patents on ML systems or inference optimization.

What We Offer

• Opportunity to work on AI and machine learning applications in real-world business contexts.
• Collaborative, innovative, and growth-driven work environment.
• Competitive compensation, benefits, and career advancement opportunities.