Senior Product Manager, Search & Exam

Qualia•Austin, TX

4d•Hybrid

About The Position

At Qualia, we've built the leading B2B real estate technology that transforms the home buying and selling experience into a simple, secure, and enjoyable process. Our SMB and Enterprise products bring together users from across the real estate ecosystem---homebuyers and sellers, lenders, title and escrow agents, and real estate agents---onto a single shared digital closing platform, providing greater clarity and transparency to real estate transactions. Today, through our business customers across the country, millions of consumers use Qualia to close on homes every year. Real estate closings run on expert judgment work — reading recorded documents, reconciling records, deciding what has to be fixed before a deal can close. It takes hours per file, the rules differ by state, and it has stayed manual for decades. Qualia Clear does this work with AI today, on real transactions. Our AI reads the documents, drafts the work product, and hands it to a professional to review and use. You'll own a defined slice end to end alongside an existing PM — and you'll own the hardest part of any AI product: knowing how good the output actually is, why it fails when it fails, and making it measurably better release over release.

Requirements

5+ years as a Product Manager building exceptional software products
Experience shipping AI or ML features where output quality was the product, not a feature of it. You've had to answer "how do you know it's working?" with data
Strong quantitative chops: fluent in SQL, comfortable in Python or a notebook, and literate in the statistics that keep you from fooling yourself — sampling, variance, significance, base rates
You've built or owned an evaluation approach — ground-truth sets, scoring rubrics, model-graded evals, regression suites — or you can clearly describe how you would
Experience building AI agents, professionally or personally. Come ready to demo one
Fluent with AI tooling (Claude Code or equivalent) as a daily working method, not a curiosity
You get what we're trying to do — how the models work, what AI changes for an industry like this — and you can get up to speed on title quickly
Comfortable working in a domain where correctness is defined by experts, not by you. You'll spend real time with practitioners learning why an answer is wrong
Hungry and proactive: you find the work, you don't wait for it
Enterprise SaaS or ERP experience; systems thinker who builds into complex, interconnected products
Exceptional written communication — tight, plain, and defensible

Nice To Haves

Knowledge of real estate, title, or escrow industries is a plus

Responsibilities

Own a slice of the roadmap end to end: scoping, release, adoption, and iteration
Own output quality as a measurable thing — define what "correct" means for AI-drafted expert work, build the evaluation and scoring approach that proves it, and defend the numbers
Build the error taxonomy: turn thousands of user corrections into a ranked, quantified list of what the model gets wrong and what fixing each is worth
Design and run experiments — prompt changes, model changes, pipeline changes — with enough statistical rigor that the result holds up
Scope AI behavior with deep, precise requirements — the specification is the product in this domain
Build rapid prototypes to pressure-test ideas before committing engineering time
Partner with Engineering and Design to break ambitious goals into small, shippable milestones
Write tight, plain-language PRDs that frame the problem, success criteria, and context
Instrument everything you ship — accuracy, adoption, latency, and cost per run — and drive cost down without giving up accuracy
Report to executives on quality trends, roadmap, and adoption
Work with GTM, CS, and customers directly to drive adoption of what you ship