AI Tooling Engineer

Qventus

36d•$120,000 - $150,000

About The Position

We're hiring Engineers dedicated to building AI-powered tools that change how work gets done inside the company. You'll embed with teams across operations, customer success, and finance, find the work that's slow and manual, and ship AI agents and automations that actually get adopted. This is a build-fast, talk-to-humans, own-the-problem role - not a research role and not a pure-infrastructure role. You'll define what "internal AI" looks like here from scratch.

Requirements

1–3 years building software — backend services, internal tools, automations, or API-driven apps. Internships and co-ops count.
Strong in TypeScript and/or Python.
You've built something with LLMs — an agent, a RAG pipeline, a copilot, a prompt-driven tool — through work, a side project, or coursework.
Comfortable wiring systems together: APIs, webhooks, SaaS integrations, async workflows.
You can talk to non-technical stakeholders, figure out what they really need, and drive adoption.
You move fast under ambiguity and iterate from real feedback.

Nice To Haves

Built copilots, agents, or workflow automations before
Used an orchestration framework (LangChain, Temporal, or similar)
Worked with vector databases or retrieval systems
Integrated with Salesforce, Slack, Zendesk, Notion, or Jira
Startup / high-growth experience

Responsibilities

Define what "internal AI" looks like here from scratch.
Partner closely with the CEO and founders on identifying the highest-leverage operational problems to solve and shaping where internal AI goes next.
Work alongside functional leaders across the business to understand their priorities, surface opportunities, and align on what to build.
Embed with Operations, Customer Success, Finance, People/Talent Acquisition, Engineering, IT and Data teams to learn their workflows and ship tools they actually use.
Build an agent that drafts customer success replies from our knowledge base and ticket history.
Build a Slack-based workflow that routes and pre-fills finance approvals.
Build an internal eval harness so we can trust the LLM features we ship.
Build a tool that lets the ops team query [internal system] in plain language.
Pick the highest-impact project, ship it, and move to the next.