Agentic AI Engineer

CMO Owls, Inc.; United States; Remoto; USD 120.000 - 180.000; Tempo pieno

Descrizione del lavoro

CMO Owls is seeking an Agentic AI Engineer to design, build, and refine specialist agents powering our marketing platform. You will shape how agents reason, what tools they call, and how they stay grounded in real customer data to deliver reliable results.

You will work with data sources like Search Console, GA4, Stripe, and other APIs, writing prompts and evaluation harnesses, choosing models, and constructing deterministic logic where appropriate.

Competenze

• 3+ years of professional software engineering experience, ideally in Python.
• Hands-on experience building production systems that call LLM APIs.
• Understanding when an LLM is the right tool and when deterministic code is better.

Mansioni

• Design, build, and maintain specialist agents across SEO, content, GEO, and social product lines.
• Write and tune prompts, tool definitions, and evaluation harnesses for each agent.
• Build MCP servers and tool integrations connecting agents to real data sources (Search Console, GA4, Stripe, APIs).
• Choose the right model and architecture for each task, favoring deterministic logic when possible.
• Build evaluation pipelines that catch regressions before customers see them.
• Work directly with customers and support to diagnose agent failures.

Conoscenze

Python

LLM APIs

Deterministic logic

Prompt engineering

Technical communication

Strumenti

FastAPI

SQLAlchemy

Descrizione del lavoro

Engineering Remote, US Full-time Mid to Senior

Design and ship the specialist agents that power CMO Owls, from technical SEO audits to GEO visibility monitoring, using real tool calling, evals, and grounded reasoning, not prompt tricks.

About this role

CMO Owls runs seventeen specialist AI agents that act as a real marketing team for our customers. As an Agentic AI Engineer, you will build and refine these agents end to end: how they reason, what tools they call, how they stay grounded in real customer data, and how they hand off work to each other. The bar is the same one frontier AI labs hold their own agent work to: does the output survive contact with a real, paying customer, or does it just look good in a demo.

What you will do

• Design, build, and maintain individual specialist agents across our SEO, content, GEO, and social product lines
• Write and tune prompts, tool definitions, and evaluation harnesses for each agent
• Build MCP servers and tool integrations connecting agents to real data sources (Search Console, GA4, Stripe, third-party APIs)
• Choose the right model and architecture for each task, favoring deterministic logic wherever a task does not genuinely need an LLM
• Build evaluation pipelines that catch regressions in agent output quality before they reach customers
• Work directly with customers and support to diagnose and fix real agent failures

What we are looking for

• 3+ years of professional software engineering experience, ideally in Python
• Real, hands-on experience building production systems that call LLM APIs, not just personal projects
• A clear understanding of when an LLM is the right tool and when deterministic code is the better answer
• Comfort reading and reasoning about prompts and agent tool definitions as real engineering artifacts
• Strong written communication, since a lot of this role is documenting how and why an agent behaves the way it does

Nice to have

• Experience with retrieval-augmented generation, MCP servers, or multi-agent orchestration frameworks
• A background in SEO, content marketing, or growth, even informally
• Experience with FastAPI and SQLAlchemy