Quality Control (QC) Lead
- Company
- Quantigo AI
- Location
- Dhaka
- Job type
- Full Time
- Salary
- Negotiable based on experience
- Experience
- 2 Years
- Deadline
- 2027-01-21
Job description
Quantigo AI is looking for a detail-obsessed VLA Quality Control (QC) Lead to spearhead quality assurance for its Vision-Language-Action (VLA) and video action captioning projects. This is not a traditional administrative or headcount-management role: most of the time is spent hands-on reviewing complex video annotations, calibrating edge cases and guiding dedicated project pods to meet strict quality standards.
Quantigo AI combines human expertise with machine efficiency to deliver annotated data at scale for clients ranging from startups and academic institutions to Fortune 500 companies. The position is full-time and permanent, based in Dhaka with a hybrid (remote/office) arrangement, and starts immediately.
Responsibilities
- Audit frame-accurate temporal segments and natural-language action descriptions against project guidelines.
- Design and run the QA workflow for VLA datasets, including review stages, sampling and acceptance criteria.
- Train, coach and assess annotators and reviewers and give feedback on performance.
- Lead and mentor assigned project teams, onboarding annotators onto new rubrics and running calibration sessions.
- Identify ambiguities in video datasets, document edge-case resolutions and feed them back to Project Managers and clients.
- Track annotator error patterns, maintain quality logs and coach to improve Inter-Annotator Agreement (IAA).
Requirements
- 2 years of hands-on experience on VLA, egocentric video captioning, temporal action localization or complex multimodal annotation projects.
- Strong understanding of how VLA models are trained and what makes demonstration data good or bad.
- Proven experience leading QA for a team and designing review processes.
- Exceptional written and analytical English, with a sharp eye for precise verbs, spatial relationships and subtle state changes in video.
- Sharp eye for physical cause-and-effect, hand-object interactions and frame-accurate temporal segmentation.
- Proven ability to dissect annotation guidelines and enforce consistency across a team.
- Strong analytical and root-cause problem-solving skills.
- Nice to have: hands-on experience with Encord, Supervisely, CVAT or similar video annotation platforms.
- Nice to have: experience building Golden Datasets, QA rubrics or training materials for AI data teams.
Benefits
- Competitive salary and festival bonuses
- Education and training assistance stipends
- Clear growth opportunities within the company
Skills
- CVAT
- Encord
- Quality Assurance
- Supervisely
- Team Leadership
- Video Annotation