Product
AI Engineer — Agent Systems & Applied AI
Why this role exists
Aurora is the intelligence layer of Finta. This role exists to design, build, and improve the AI systems that power autonomous deal workflows — research, drafting, enrichment, reasoning, and automation across the Finta workspace.
The AI Engineer ensures Aurora becomes a true capital copilot capable of understanding deal context, recommending next actions, and executing workflow steps for users.
Captain / Co-pilot ownership
- Captain (primary): AI capability roadmap and reliability of Aurora systems.
- Co-pilot (supports): Growth, Sales, and Success teams by building AI tools that accelerate their workflows (research agents, email drafting, deal insights, etc.).
- Non-negotiable: AI systems must produce observable outcomes tied to user workflows and stage progression.
What success looks like
30 days
- Understand the current Aurora architecture and model orchestration stack.
- Identify top 3 reliability or capability improvements and ship the first upgrades.
90 days
- Ship at least one major AI workflow improvement (e.g., research agent, automated contact enrichment, email reply suggestions).
- Improve latency, accuracy, or cost efficiency of key AI calls.
6 months
- Aurora meaningfully reduces manual work for users by automating multiple deal workflow steps.
- AI capabilities measurably improve activation, expansion readiness, or advisor productivity.
Tools you'll likely use
LLM APIs (OpenAI / Anthropic or similar), LangGraph / LangChain-style orchestration, vector databases, Python / TypeScript services, and internal workflow tooling.
Reports to: CEO / Head of Product
Responsibilities
- Design and implement AI agents that perform workflow tasks (research, summarization, drafting, classification). - Build orchestration logic connecting models, APIs, and internal tools. - Improve prompt systems, evaluation pipelines, and reliability safeguards. - Work with product to design AI-native features inside the Finta workspace. - Integrate external data providers (enrichment, research, CRM signals). - Optimize performance and cost of LLM usage. - Build evaluation metrics so AI output quality can be measured and improved.
Requirements
- Strong software engineering ability (backend and system design). - Experience building applications using modern LLM APIs or open-source models. - Ability to design agent workflows and tool-calling architectures. - Comfort working in ambiguous, fast-moving product environments.
Nice to have
- Experience with LangChain, LangGraph, or similar orchestration frameworks. - Experience with vector databases, embeddings, and retrieval systems. - Experience building AI copilots or workflow automation systems.