Description
Summary We are building an AI workforce platform: autonomous AI "employees" that run operations for small businesses. Not chatbots or copilots, but agents that execute real work end to end (answering, quoting, invoicing, reconciling, reporting) under owner-defined policies, either fully autonomously or alongside the owner. We are funded, the architecture is specced, and the first agent is "Sam," an Operations and Finance agent. The framework decision is down to two candidate stacks, and the first milestone of this engagement IS the evaluation. Milestone 1 is a fully paid, fixed-price milestone funded through Upwork before work begins, and all communication stays on Upwork. You will stand Sam up on Hermes in week one, run it unattended, and produce the evidence report that decides the production foundation. Then you build production on the winning stack. We are looking for someone who has shipped autonomous agent systems and can evaluate a framework honestly, not someone who arrives with a religion.