TikTok's 250 Nashville Cuts: A Trust and Safety Resume Rewrite

TikTok filed its WARN notice on August 5, 2026. The Nashville office at 827 19th Avenue South closes October 5, 2026, taking 250 trust and safety and content moderation jobs with it. Severance runs one month plus one additional month per year served, so for most of the cohort the realistic resume-to-offer window is roughly eight weeks on payroll, from August into November.

The problem is not experience. Nashville alumni ran policy enforcement at the scale of a billion-user platform. The problem is translation: NDAs, ticket-queue metrics that mean nothing outside TikTok, and a job title ("Content Moderator") that reads to an Anthropic or Discord recruiter like vendor queue work. This is the rewrite.

Why "Content Moderator" is now a career trap

The title that got you hired at TikTok is the title that will keep you out of the AI labs that are hiring. In Refolk's index of professional profiles, 925 US professionals currently carry a "Trust and Safety" or "Content Moderator" title. Only 3 use the keyword cluster "trust safety policy AI red team" together, and those three sit at Google, the Integrity Institute, and Character.ai.

That is a near-empty keyword category. The 250 Nashville alumni are large enough to double or triple it overnight if they rebrand before the ATS reads them. The mechanism is boring: AI-lab recruiters at Anthropic Safeguards, OpenAI, and Character.ai search for "Policy Enforcement," "Safeguards Analyst," "Integrity Operations," and "Red Team." They do not search for "Content Moderator," because the title is dominated by BPO contractors they have already filtered out.

Here is what the index actually looks like at the employer level:

Cognizant, Accenture, and Wipro hold the top employer slots. Platform-side employers like TrustLab, Fizz, Tools for Humanity, and Quizlet appear, but they are a minority. For a platform-side trust and safety resume, that is leverage, not a problem. A TikTok badge is first-party work at a platform that regulators actually name. Meta, Discord, and Roblox hiring managers explicitly prefer first-party operations experience over vendor queue tenure.

Rename the role before the ATS reads it

Do this on the title line, in the summary, and in at least two bullets per role:

• "Content Moderator" becomes "Policy Enforcement Specialist" or "Integrity Operations Analyst"
• "Trust and Safety Agent" becomes "Safeguards Analyst" or "Policy Enforcement Lead"
• "Quality Analyst" on a moderation queue becomes "Policy Calibration Analyst"
• "SME" or "Escalations" becomes "Appeals Reviewer" or "Senior Enforcement Reviewer"
• "Wellness Captain" or peer-support rotation becomes "Trauma-Informed Operations Program Lead"

Keep the TikTok title on the role itself for ATS parsing of the employer. Reframe the function in the bullets. This is the exact friction Refolk is built for: paste the Anthropic, Discord, or Roblox posting, and the resume comes back retitled and rephrased against that specific job's language rather than TikTok's internal taxonomy.

Speak in millions, not tickets per shift

AI-lab trust and safety is measured in enforcement volume, appeal rate, and overturn rate, not tickets per hour. Anthropic's transparency reporting lists 11.4 million banned accounts in H1 2026, 398,000 appeals, and 42,000 appeal overturns, an overturn rate of roughly 10.6 percent. That is the shape of the number a hiring manager reads first.

A Nashville moderator who writes "resolved 320 tickets per shift at 96% QA" is speaking BPO. The same person who writes "actioned enforcement decisions contributing to a policy area averaging X million monthly removals, with a 7% appeal rate and a 4% overturn rate" is speaking Anthropic. The underlying work is identical. The frame is not.

Numbers Nashville alumni can translate without breaking NDA

Most TikTok operational metrics are confidential. These are not:

1. Monthly action volume for your policy area, aggregated and rounded, framed as "a policy area averaging..." rather than a specific internal figure
2. Appeal rate on your queue, expressed as a percentage
3. Overturn rate on appeals you reviewed, which demonstrates calibration quality
4. Policy write or policy update contributions, counted as artifacts shipped, not as drafts
5. Calibration or QA scores at the team level, framed as "sustained team QA at X%"
6. Wellness or resilience program participation, counted as cohort size and cadence

If a figure is covered by NDA, drop the number and keep the verb. "Led calibration for a high-severity policy area" is still better than "Processed tickets."

The target employers are hiring under five different names

The AI labs, the platforms, and the policy nonprofits all want the same person. They just do not advertise under the same title. If you only search "content moderator" or "trust and safety," you will miss most of the live roles.

• Anthropic Safeguards publishes transparency numbers (11.4M banned accounts, 398k appeals in H1 2026) and hires appeals reviewers, policy enforcement, and red team
• OpenAI is actively hiring political engagements and policy roles per CNBC reporting
• Discord hires "Safety Operations" and "Policy" rather than "Moderator"
• Roblox hires "Civility" and "Safety Engineering" at scale
• Meta hires "Integrity" across Facebook, Instagram, and WhatsApp
• Character.ai and Tools for Humanity appear in Refolk's index as platform-side T&S employers for people who prefer startup pace
• TrustLab is the independent-vendor option for people who want policy craft without BPO queue work

The European tailwind matters here. Under DSA and AI Act powers, the European Commission can demand to inspect models, restrict EU market access, and fine providers up to 15 million euros or 3% of annual turnover. Every US AI lab shipping into the EU now needs DSA-literate enforcement staff. Nashville alumni who worked inside TikTok's DSA compliance workflows are strategically rare, and that line belongs in the summary, not buried in a bullet.

Trauma is a credential when you frame it as program work

Burnout and exposure to graphic content are not disqualifications. They are the inside knowledge AI labs are quietly hiring for. The Integrity Institute's "Trust in Tech" podcast, hosted by Alice Hunsberger (VP of Trust and Safety and Content Moderation at PartnerHero), runs full episodes on trauma-informed workflows with T&S veterans like Cathryn Weems. The canon treats "more holistic, trauma-informed approaches to building resilience and better working conditions" as a professional specialty, not an HR disclosure.

What that looks like on a resume:

• "Designed and ran peer-support rotations for a 40-person policy team, including graphic-content exposure caps and post-shift debrief protocols"
• "Co-authored wellness SLA adopted across three regional teams"
• "Trained new reviewers on trauma-informed escalation, reducing median exposure to high-severity content by X%"

Program verbs ("designed," "ran," "trained," "co-authored"), cohort sizes, and durations. No disclosure of personal impact. This is wellness operations work, and it reads as program management, which is exactly the altitude Meta Integrity and Anthropic Safeguards are hiring for.

Three credentials to add in the next two weeks

Join the Trust and Safety Professional Association, apply to the Integrity Institute, and list both on the resume above your education section. Both are specifically designed to help platform-side T&S workers find each other and get hired.

• Trust and Safety Professional Association (TSPA) has roughly 10,825 LinkedIn followers. It is a 501(c)(6) industry body with membership, events, and a job board that AI-lab recruiters actively read. Membership is resume-ready the day you join.
• Integrity Institute admits applicants with at least six months of professional integrity experience. Nashville alumni with any TikTok T&S tenure clear that bar. The Institute runs clinics, the "Trust in Tech" podcast, and a referral network that disproportionately feeds AI-lab Safeguards teams.
• Public writing on policy craft, even one post on your own site about calibration or appeals mechanics, becomes a portfolio line. Anthropic and Character.ai read these.

Put TSPA and Integrity Institute in a "Professional Memberships" section between experience and education. Both signal that you treat the field as a profession rather than a job, which is the exact signal that gets a Safeguards recruiter to click your LinkedIn.

The eight-week sequencing plan

With a WARN date of October 5 and severance that runs one month plus one per year served, most of the cohort has an August-to-November window to land. Do this in order:

1. Weeks 1-2 (August): Rename every title on the resume. Join TSPA. Apply to Integrity Institute. Draft the summary around DSA literacy and policy enforcement at platform scale.
2. Weeks 3-4: Translate every ticket metric into monthly action volume, appeal rate, overturn rate. Rewrite three bullets per role. Draft a trauma-informed program line.
3. Weeks 5-6: Apply to Anthropic Safeguards, OpenAI policy enforcement, Discord Safety, Roblox Civility, Meta Integrity, Character.ai, TrustLab, and Tools for Humanity. Tailor the cover letter per posting. This is the step that eats the week if you do it by hand, and the step Refolk collapses to minutes by scoring your fit against the posting and drafting the letter from your own history.
4. Weeks 7-8 (October): Second-wave applications to policy nonprofits and vendor-adjacent roles. Reach out to the Integrity Institute referral channels. Line up references from inside TikTok before badges deactivate.

The Nashville cohort is competing with a global TikTok T&S diaspora. There was a February 2025 global restructuring and a July 1, 2026 wave across Dublin, Singapore, and Indonesia. Three rounds in 18 months means the AI labs have seen a lot of TikTok resumes already. The ones that get interviews read like Anthropic's transparency report, not like a BPO performance review.

FAQ

Can I list TikTok on my resume if I signed an NDA?

Yes. Employment history, dates, title, team, and the general scope of your policy area are not covered by a standard NDA. What is covered is specific internal metrics, unreleased product details, named user data, and internal policy documents. Keep the employer, the title, and the function. Rephrase quantitative claims as rates, percentages, or aggregated monthly volumes rather than internal dashboard numbers, and omit anything you would not say publicly on LinkedIn.

Is "Content Moderator" really worse than "Policy Enforcement Specialist"?

For AI-lab recruiting, yes, and the mechanism is ATS keyword scarcity. In Refolk's index, 925 US pros use the legacy title and only 3 use the "AI red team / policy / T&S" keyword cluster. Recruiters at Anthropic, OpenAI, and Character.ai search the sparse category. The underlying work does not change. The retrieval surface does.

Should I apply to BPO roles at Cognizant or Accenture as a backup?

As a bridge, not a destination. Cognizant and Accenture are the top T&S employers in the index by headcount, so the roles exist and pay quickly. But platform-side experience is the differentiator that moves you toward Meta, Discord, or Anthropic, and taking a BPO queue role for 12 months can make the next pivot harder, not easier. If you need the income, take it and keep applying platform-side in parallel.

How do I write about graphic-content exposure without disclosing trauma?

Frame it as program work. "Designed peer-support rotations," "trained reviewers on trauma-informed escalation," "co-authored wellness SLA." Cohort sizes, cadence, and policies, not personal impact. The Integrity Institute canon treats trauma-informed workflow design as a professional specialty, so the vocabulary already exists. Use it.