The signal-to-noise crisis isn’t just a recruiting problem — it’s your biggest cross-border blind spot
When you hire a remote sourcing agent in Shenzhen, a media buyer in Manila, or a full-service agency in London, you are making a bet on reputation. Your standard due diligence is a LinkedIn profile, a handful of recommendation blurbs that read like wedding toasts, and maybe a Zoom call where the other party has rehearsed answers to your top three questions. The bet usually pays off 60% of the time — and that 40% of misses costs you months of wasted ad spend, inventory write-offs, or compliance fines. What if you could verify a person’s actual working style — how they communicate, whether they deliver on time, how they handle conflict — before you wire a deposit? That is the promise of a little tool called Badge, and even though its makers are targeting hiring managers for software engineers, the mechanism is far more relevant to cross-border e-commerce operations than to any Silicon Valley talent team. Let me walk through why.
What Badge actually solves (and why your supply chain needs it)
Badge uses AI agents to collect anonymous, work-email-verified peer reviews and distill them into a portable Trust Score. The creator, Lokesh Motwani, spent 15 years hiring engineers and realized that resumes are increasingly useless because “AI can generate polished resumes, cover letters, and LinkedIn profiles in minutes.” The core innovation is not the aggregator — it’s the anonymity plus the AI-guided interview process that digs into how someone works, not just what they claim to have done.
For a cross-border seller, substitute “engineer” with “supplier account manager,” “freelance copywriter,” or “logistics partner.” Every day you rely on people whose actual performance is buried in internal Slack threads, 360° feedback systems, or private WhatsApp groups that you can never access. Badge’s model extracts that signal and makes it portable. The reviews are verified because the reviewer must use a work email from the same organization, and the AI agent prompts with role-specific questions — for a software engineer it asks about code quality; for a sourcing agent it could ask about lead time management, defect rate escalation, and negotiation style. The product is not currently tailored to e-commerce roles, but the architecture is trivial to adapt.
The anonymous honesty edge
LinkedIn recommendations are a joke. They are reciprocal back-scratching — you write me a glowing paragraph, I write you one. No one admits, “This person was great but sometimes missed deadlines when under pressure.” Badge’s anonymous format eliminates the social pressure to inflate. In the Product Hunt comments, one user, Buse, remarked that the overnight Trust Score felt “more honest” than any LinkedIn recommendation she’d received. That honesty is exactly what you need when assessing a potential partner in a jurisdiction where legal recourse is expensive or impractical.
How it differs from the incumbents you already lean on
Most cross-border operators use a patchwork of tools to vet talent:
LinkedIn: The recommendations are public, non-verifiable, and curated by the profile owner. Badge’s AI agent, by contrast, independently pulls contacts from your address book and reaches out without the candidate cherry-picking who gets asked. The maker Lokesh clarified that “the candidate has no control over who gets asked” because the agent scans contacts and identifies colleagues.
Upwork / Fiverr: These platforms have ratings, but they are easily gamed — buyers leave vague praise to maintain relationships, and sellers can pad profiles with fake projects. Badge’s work-email verification (same organization) and anonymous format make it harder to game. You can’t ask your cousin in a different company to vouch for your agency work.
Traditional reference checks: Every seller has called a reference and heard vague platitudes. Badge’s AI agent guides the reviewer through structured questions — communication style, collaboration, ownership, reliability — and compiles a scored report. The creator emphasized that “the agents will ask mix of both actual work quality and interpersonal skills questions … the work quality questions depend on the role.”
Why Amazon sellers should care more than Shopify ones
Amazon seller central is a high-stakes environment. One bad supplier can tank your listing with poor unit quality, delayed shipments that ruin your IPI score, or compliance violations that get your account suspended. Shopify brands, while also reliant on partners, have more flexibility to swap suppliers or agencies because they control the storefront and fulfillment can be more modular. For Amazon sellers, a sourcing agent who lies about factory audits or a prep center that mislabels units can cause irreparable damage to a brand’s BSR trajectory. Badge’s portability is critical here — if a sourcing agent has a verified Trust Score from previous clients, you can see patterns: did they reliably communicate during peak seasons? Did they own up to mistakes? That is the kind of signal you cannot get from a PDF portfolio.
Where the math breaks
Badge is not a silver bullet. The system depends on network effects — you need enough reviewers in the same organization to build statistical significance. The creator acknowledged that outliers matter less with a sample size of 20–30 reviews, but most cross-border relationships involve teams of 2–10 people. A freelancer who had only 3 coworkers may end up with a Trust Score that reflects those 3 people’s biases. Additionally, the AI agent draws contacts from the user’s address book, but on iOS the user can “select few” — a degree of curation remains. The makers claim that “anyone can still review you using LinkedIn and phone number,” but that introduces noise from people who never worked directly with the candidate.
The biggest gap for cross-border use: Badge verifies that two people worked in the same organization, but it cannot verify the quality of that organization’s standards. A glowing review from a fly-by-night agency might say more about the agency’s low bar than the individual’s skill. The Trust Score is relative to the peer group, and in e-commerce, peer groups can be notoriously insular.
What cross-border sellers can borrow from this model
Even if Badge never ships a “Supplier Profile” feature, the underlying pattern is something you can implement inside your own operations today. Here is how I would adapt the concept for a DTC or Amazon brand:
Internal peer reviews for remote teams: If you have a distributed team of VA’s, customer support agents, and sourcing managers, run an anonymous pulse survey three months in — not for performance review, but for a “collaboration score” that you can share with future employers or agencies. It builds a portable reputation within your niche.
Supplier vetting via peer networks: Instead of asking a supplier for references (they will give you their best customers), ask them to invite their previous clients through a tool like Badge. You get anonymous reviews from people who actually bought from that factory, not the cherry-picked testimonial page.
Agency hiring with a trust score gate: When interviewing a media buying agency, ask each of their team members (not just the founder) to share a Badge profile. If the account managers have low Trust Scores it tells you more about the agency culture than any pitch deck.
Two specific tests to run this week
Shadow test with a freelancer you already trust: Ask a VA or designer who has worked with you for six months to create a Badge profile (it’s free for the current beta). Connect your LinkedIn and see if the AI agent identifies colleagues from your shared projects. The feedback you get back could serve as a baseline for how honest the tool really is. Spend 30 minutes reading the generated Trust Score — does it match your gut feeling about that person’s communication style and reliability? If yes, you have validated the concept for your own hiring funnel.
Recruiter-side trial: Use the recruiter portal to check a candidate you are considering for a role — for instance, a logistics coordinator who claims experience with large-scale China-to-US fulfillment. Paste their email or LinkedIn URL. If they have a Badge profile, great. If not, ask them to get one — their willingness to do so is itself a signal. Early adopters of Badge will be people who value transparency. That is exactly the kind of partner you want in cross-border e-commerce.
My judgment: an early-stage signal worth monitoring
Badge is a three-person bootstrapped startup — 1 engineer and 2 interns. The product is clearly in alpha. The anonymous review concept is well intentioned, but the trust model depends on the AI agent’s ability to parse nuance, and the sample size problem is real. For cross-border sellers, the immediate use case is not supplier vetting (too few reviewers in most supply chains) but internal team reputation and agency account manager screening.
That said, the problem Badge tackles — the erosion of trust in credentialing systems — is exactly the pain point every e-commerce operator feels when they onboard a new partner across time zones and legal jurisdictions. If Badge gains traction in the professional hiring market, the same framework will inevitably be adapted for B2B service marketplaces. I will be watching how they handle the selection bias issue (Gal Dayan’s trenchant question about whether the AI agent truly prevents cherry-picking) and whether the Trust Score can ever account for the fact that a 5-star peer review in a low-stakes environment does not equal a 5-star review in a high-stakes Amazon FBA operation.
For now, the smart move is to treat Badge as a beta tool for a single high-stakes hire — a sourcing manager, a PPC agency lead, or a logistics coordinator — and benchmark the results against your own gut check. If the signal aligns, you have found a new arrow in your vetting quiver. If it doesn’t, you have lost only 20 minutes of clicking. Either way, you will be ahead of 90% of sellers who are still making decisions based on PDFs and handshake vibes.
