KlearNow.AI is an AI-native customs brokerage and trade compliance platform transforming how global enterprises clear goods and manage duty exposure. We serve major importers and logistics providers including Target, BASF, Hershey, and SLB, by replacing manual entry work and brittle rules engines with an autonomous ingestion and clearance pipeline. Our north-star metric is Zero Touch Rate: the share of shipments that clear end to end with no human intervention.

We're hiring a Product Manager to own the Customs House Brokerage (CHB) product area, the core of what we do. You'll own the products that classify goods, build and validate entries, file to ABI/ACE, handle PGA and tariff logic, and get shipments across the border correctly and automatically.

This is not a generalist SaaS role. You need real broker depth; enough to know when the system is right, when it's wrong, and when it's confidently wrong, because a plausible-looking error here is a seven-figure duty discrepancy waiting to happen. And you must be AI-native: someone who uses AI as their primary means of production and designs products that are agentic by default.

You'll own CHB end to end, from problem framing to shipped, measured outcomes. That spans classification (vector-DB retrieval with LLM fallback, citation and confidence scoring), entry construction and validation, ABI/ACE filing logic, PGA handling, and the tariff stack beneath it all: Section 232, Section 301, and the IEEPA/Section 122 layer including Chapter 99 stacking.

Every manual touch our ops team makes is a defect in the product, not a fact of life. You'll find the exceptions eating human time and drive them out one workflow at a time, writing acceptance criteria in terms of what the system does with no human action required: the trigger, the automated behavior, the pass criteria, the failure modes. You'll co-own delivery with an engineering lead inside our Bet / Shape / Build / Validate pod framework, and ship prototypes as part of your thinking, not after it.

Customs brokerage depth. You know the entry lifecycle cold: HTS classification, 7501, ISF, ABI/ACE/CATAIR mechanics, PGA, and post-entry work (PSC, PEA). You can hold a real conversation about 232/301, the current IEEPA/Section 122 landscape, and why the order of operations in a duty calc is not negotiable. A US broker license (or the field-earned equivalent) is a strong plus.

Technical fluency. You understand how data moves through a platform, ingestion, extraction, APIs, integrations, and the difference between a deterministic rule and a probabilistic model. You can read a schema, reason about where a bug lives, and write requirements engineers respect. You prototype and query data yourself.

Product craft. You've owned products end to end in a complex, high-stakes domain. You do real discovery, prioritize ruthlessly against a metric that matters, write crisp testable criteria, and ship on a cadence at the quality bar Target and BASF expect.

This part isn't negotiable. The bottleneck has shifted from producing artifacts to having the judgment to know what's right; and AI collapses the first 70% from weeks to hours. We expect you to work at that speed: use Claude to frame problems and pressure-test your reasoning, use Cursor / Lovable / v0 to get a working prototype in front of people in a session, and think in agentic terms because systems that retrieve, reason, cite, score their own confidence, and escalate only true exceptions are our product. We're not looking for someone who “has used ChatGPT.” We want someone whose default working mode already assumes these tools exist.