> Markdown version of [/jobs/ext/4167235-ai-engineer-support-automation](https://www.wearedevelopers.com/jobs/ext/4167235-ai-engineer-support-automation). Every page supports `.md` or `Accept: text/markdown`. Links point to the HTML versions so they work for humans too. Agent guide: [/agents.md](https://www.wearedevelopers.com/agents.md).

---

# AI Engineer, Support Automation

- **Company:** PointFive LLC
- **Location:** New York, NY, United States
- **Salary:** $120,000 - $195,000
- **Contract:** Permanent contract
- **Skills:** Application Programming Interfaces (APIs), Artificial Intelligence, Amazon Web Services, Microsoft Azure, Cloud Computing, Software Debugging, Systems Integration, Pylon Synchronization Software, Large Language Models, Agentic-AI, Build Management, Zendesk, Webhooks
- **Published:** October 31, 2026
- **Apply:** https://www.indeed.com/viewjob?jk=d1356a8a3fcb7e41

## About the Role

Must-have

 * Shipped AI agents or LLM-powered tools with real users.
 * Hands-on experience integrating and debugging APIs, webhooks, and data sources.
 * Built evals to measure agent quality before and after shipping.
 * Clear, concise writing for both customers and engineers.
 * Ability to operate in a startup environment: adaptable, self-directed, and resourceful.

Nice to have

 * Cloud or FinOps knowledge (AWS, Azure or GCP billing)
 * Customer-facing technical experience (support, solutions, or forward deployed engineering)
 * Hands-on experience with AI support tooling (Pylon, Intercom Fin, Zendesk AI or similar)

## Description

As PointFive's first AI Engineer, Support Automation, you'll own technical support for our enterprise customers. You'll design and build an agentic solution to triage, diagnose, and resolve customer issues end to end. You'll also measure how well it's working, and own the escalation path to Engineering for anything that needs a fix.

This is an individual contributor role with ownership of an entire function, based in our NYC office (hybrid).

What You'll Do

 * Design and build an agentic solution to resolve customer tickets end to end, from first response through diagnosis and resolution.
 * Define what success looks like and build the measurement layer to track it (time to first response, time to resolution, CSAT).
 * Work the escalations your agents can't solve yet, then teach the system to handle them next time.
 * Partner with Engineering and Product on recurring issues, routing confirmed bugs to Engineering's coding agents with the context needed to fix them.
 * Turn resolved tickets into self-serve documentation customers can use on their own.
 * Feed support trends back into the product roadmap.