Product Analyst, Afficiency

• Candidates not currently living in the NY tri-state area will not be considered
• About 2+ years of business, analyst, operations, or product-adjacent work experience — a non-technical background is completely fine
• A genuinely analytical mind: you notice when something doesn't add up, and you dig until you understand why
• Strong common sense and business judgment about how a product or process should work, and an eye for where it doesn't
• A self-starter attitude — you don't wait to be told what to look at or handed a fully scoped project
• Real ownership instinct: when you find a problem, your instinct is "let me go fix this," not "let me go tell someone."
• Comfort learning and using AI tools (like Claude) as a working tool to move faster on analysis and execution — no coding background required, just curiosity and a willingness to figure it out
• Clear written and verbal communication — you can explain what's broken and why it matters to both technical and non-technical people

Desirable

• This is a hybrid analyst/builder role
• You'll spend part of your time in our products and UX flows like a detective — reading session recordings, walking through application journeys yourself, digging into support tickets and drop-off points — to find where users get confused, stuck, or frustrated
• Then, instead of stopping at a slide deck of recommendations, you'll use AI tools (Claude and others) to prototype and ship the fix yourself, or work directly with the team that can
• A year in, success looks like a track record of real, shipped improvements you identified, scoped, and drove to completion — not a backlog of tickets someone else implemented
• Own the ongoing review of Afficiency's product and UX/UI journeys — application flows, agent tools, and consumer-facing experiences — to identify friction, confusion, and drop-off points
• Dig into real usage data: session recordings, support tickets, underwriting outcomes, and conversion funnels, to find where the product isn't working the way it should
• Turn findings into concrete, prioritized recommendations grounded in business impact, not just "this feels off."
• Use AI tools such as Claude to prototype, build, or directly implement fixes — copy changes, flow adjustments, small tooling, automation, or reports — rather than only handing off findings
• Partner with product, engineering, and underwriting stakeholders to scope and ship changes, escalating and collaborating when something needs deeper technical work
• Build a repeatable habit of measuring whether a change actually worked, and iterating when it didn't
• Bring plain business judgment to the table: does this make sense, would a real customer understand this, is this worth fixing now or later