AI-Driven Delivery Systems Engineer

AspenView Technology Partners, Inc.; United States; Remote; USD 180,000 - 240,000; Full time

Job summary

AspenView Technology Partners is seeking an Agentic Delivery Systems Engineer to own how agentic software delivery works on a large AI-led modernization program for a global payments/fintech platform. You’ll own context, token economics, tools, and permissions to ensure faster delivery without compromising quality.

You will design and optimize context handling, cost models, and model selection across frontier and smaller reasoning models, while integrating with OpenAI, Codex, Copilot, and

Qualifications

• 5+ years building and shipping production software.
• Travel or hospitality domain experience is mandatory.
• Hands-on production experience with OpenAI and agent frameworks.

Responsibilities

• Design what enters the context window: retrieval strategy, scoping, summaries, pruning rules.
• Diagnose and fix context degradation and recall loss.
• Optimize prompt caching, session compaction, and reuse to reduce token spend.
• Build cost models for agentic work and forecast for cost-reduction programs.
• Choose and size models per task tier; reserve capacity for hard reasoning.
• Own tool surface design and repository-level configuration as versioned artifacts.
• Build integrations with client systems (Azure DevOps, Repos, CI, data platforms).
• Maintain governance and security for agentic workflows; enforce PCI/DSS-like controls.

Skills

Python

TypeScript

CI/CD

English (C1)

Analytical thinking

Education

Bachelor’s degree in Computer Science or related field

Tools

OpenAI APIs

Codex toolchain

GitHub Copilot

Azure DevOps

Job description

AspenView Technology Partners is seeking an Agentic Delivery Systems Engineer to own how agentic software delivery works on a large AI-led modernization program for a global payments/fintech platform. You’ll own context, token economics, tools, and permissions to ensure faster delivery without compromising quality.

You will design and optimize context handling, cost models, and model selection across frontier and smaller reasoning models, while integrating with OpenAI, Codex, Copilot, and