Annotation Operations Manager — Vietnam

Location: Remote Contract: Full-time, with founding-team equity Reports to: Chief of Staff / Data Operations Lead

Our mission

We're building the AI data infrastructure that makes Vietnamese-language models world-class. Every dataset we deliver makes Vietnamese AI stronger.

We believe Vietnamese AI deserves world-class infrastructure — and deserves to be built by Vietnamese people, for Vietnamese people. We want Vietnamese AI to lead, not follow other markets.

Tinh Lọc turns Vietnam's media archives (audio, video, documents, broadcasts — historical and modern) into verified, structured, model-ready datasets. Every dataset we ship expands Vietnamese AI capability across the industry.

We have an anchor customer with a multi-year scope of work, and we launch operations within the next 30 days. This is a chance to join at the founding-team stage, not into a settled company.

Why this role matters

At Tinh Lọc, operations IS the product — and the annotator workforce is the engine that produces it. Our AI pipeline produces a first-pass extraction; your team reviews and corrects that output, then structures it into the verified, model-ready datasets customers pay for — human-in-the-loop correction at its core. This is skilled work, not commodity labour — the difference between a raw transcript and a model-ready dataset is the judgement your annotators apply on every task.

This is a true greenfield — and that's the appeal. You design the training programme, the quality system, and the L1→L2 career ladder from first principles, the way you've always believed they should be built, with no inherited broken processes to unwind. And you're not starting cold: a working AI pipeline already feeds your team clean first-pass output, a funded anchor customer gives you real, paid work from week one, and you build alongside the CEO, the Data Operations Lead, and a Senior Annotator + Trainer. You start with a small founding team of annotators and grow it to 30-50 as we scale — and the operation you design becomes the playbook the whole company runs on.

This is an ownership role on the founding team, not a support role. You own hiring, the 8-stage training programme, the quality systems, and the L1→L2 career ladder. Your operational decisions directly shape Tinh Lọc's throughput, accuracy, and unit economics.

Who we're looking for

You are someone who:

• Has 5+ years managing teams of 10-50 people in data labelling, BPO, content operations, contact centre, or a junior dev shop. You've hired at volume, trained new joiners systematically, and held a team to measurable quality targets — and you've done it with a Vietnamese-native workforce or one very much like it.

• Can design annotation workflows in Label Studio — XML annotation-interface configuration for Vietnamese transcript correction, OCR cleanup, entity tagging, and taxonomy application. If you haven't used Label Studio specifically, you can show us a comparable tool you mastered fast and explain how you'd come up to speed in weeks, not months.

• Builds structured training programmes and promotion ladders. You've designed onboarding that takes a raw new joiner to full production reliably, and you've run an explicit promotion path (e.g. L1→L2 gated on >95% accuracy + <5% reject rate). You think in stages, gates, and measurable criteria — not vibes.

• Runs real quality-control systems. Gold-task seeding, reviewer sampling, calibration meetings, inter-annotator agreement (kappa). You know that accuracy is something you engineer into the workflow, not something you hope for.

• Schedules a workforce against SLAs — shifts, holidays, sick leave, capacity planning at moderate scale, with the turnaround and accuracy targets always in view.

• Can manage a Vietnamese workforce. You understand Vietnamese working culture — how to plan Tết / Lunar New Year coverage, how to handle family obligations with sensitivity, how to recruit junior annotators through Vietnamese university partnerships, and how to balance hierarchy norms with the need for annotators to escalate edge cases rather than guess.

• Native or near-native Vietnamese; working-fluent English. The workforce is Vietnamese-native and all instruction is in Vietnamese; you'll also coordinate with the Chief of Staff and the wider team in English. Both languages need to be natural.

Who we're NOT looking for

To save your time and ours, here are profiles that won't fit:

• People who have only managed Western workforces. Managing a Vietnamese-native annotation team is a different discipline — the culture, communication norms, and retention levers don't transfer cleanly. This will surface in the interview.
• People who treat annotation as commodity labour. If you see annotators as interchangeable headcount to be churned, you'll produce poor accuracy and high turnover. We correct AI output to a quality bar; that needs skilled, retained, respected operators.
• People with no quality-control background. If you've never run gold tasks, reviewer sampling, or calibration meetings, you don't yet have the toolkit this role requires. Accuracy at scale is built on these mechanisms.
• People who want a settled org with ready-made processes. This is a build-from-zero, founding-team role. If you need an existing playbook to operate inside, this isn't the role for you.

What you'll do

First 2 weeks

• Absorb the business plan (worker architecture, tech stack, QA system); shadow the Pipeline / AI Engineer to see exactly what your team will be correcting
• Deploy a Label Studio production instance and configure the first Vietnamese transcript-correction workflow

By Day 30 — first sample batch through correction

• Run the first sample batch through correction and structuring — capture throughput, accuracy, and reject-rate baselines
• Recruit the first 2-3 annotators (university partnerships + LinkedIn Vietnam) and draft the 8-stage training programme
• Co-hire the Senior Annotator + Trainer with the Data Operations Lead

Day 30-90 — pilot + scale the team

• Support delivery of the first full pilot batch to the customer (~Day 60)
• Grow the annotation team to 5-7; stand up the QA system with the QA Lead (gold tasks, sampling cadence, calibration, inter-annotator agreement)
• Ship the Annotation Operations Runbook (daily workflows, gold-task design, escalation paths)

6-12 months

• Run the L1→L2 promotion ladder; monthly recruitment cycles and deeper university partnerships
• Build annotator retention and culture (Vietnamese workplace norms); continuously tighten workflows and capacity planning as the team grows toward 30-50

Compensation

How to apply

Send via email to contact@coreywilton.org (or through the advisor who introduced you to this role):

1. CV — concise, focused on workforce-management and data-ops experience
2. A short note (200-400 words, Vietnamese or English) answering these 3 questions:
   • Describe a labelling or data-ops team you scaled — its size, how you trained new joiners, and how you measured and maintained quality (gold tasks, inter-annotator agreement, calibration).
   • Our annotators correct AI first-pass output rather than label from scratch. Design the workflow, the QA gates, and the L1→L2 ladder you'd use to make the output model-ready.
   • Recruiting and retaining 30-50 remote Vietnamese annotators is hard. Give us a concrete university-recruitment plan, plus the biggest people-or-quality risk you foresee and how you'd pre-empt it.

We'll respond within 5 business days to all qualified candidates. Round-1 interview with the CEO (60 min, video). Round-2 with the CEO + co-founder (90 min, video). Offer extended within 7 days of Round-2 if there's a fit.