Job Highlights

• laptop to be provided
  
  

Role Summary

1

Responsibilities

Prompt Validation & Data Annotation

• Execute Multi-Step GenAI Tasks: Accurate interpretation and systematic deployment of complex generative sequences (e.g., classify\( ightarrow\) reason\( ightarrow \) respond).
• Interpret Ambiguity: Decode loosely defined client guidelines or vague initial guidelines into clear, functional instructions for model alignment.
• Labeling Taxonomy Application: Categorize complex data streams systematically, establishing pattern recognition across massive datasets.
  
  

AI Reasoning, Safety & Quality Assurance

• Detect Model Vulnerabilities: Identify edge cases, logical inconsistencies, and factual hallucinations within model outputs.
• Rubric-Based Evaluation: Grade and critique complex AI outputs against detailed qualitative rubrics for accuracy, relevance, and safety.
• Fact-Checking & Synthesis: Perform deep primary research to validate data claims, compressing findings into highly structured and usable training datasets.
  
  

Iterative Refinement & Collaboration

• Probing & Optimization: Engineer logical follow-up prompts and iterative inquiries to test the boundaries of and maximize model capability.
• Instructional Communication: Translate highly technical or abstract model behaviors into structured, actionable feedback loops for development pipelines.
• Human-in-the-Loop Pipeline: Collaborate closely with engineering and shared data teams to implement policy and model updates.
  
  

Qualifications

Linguistic & Writing Command

• Advanced Writing Proficiency: C1-level written capability or equivalent mastery, displaying clean text syntax, structural accuracy, and stylistic adaptability.
• Creative & Structured Text Design: Experience drafting diverse copy styles, ranging from imaginative narrative tasks to highly rigid, formatted technical instructions.
  
  

Cognitive & Analytical Abilities

• High Reading Comprehension: Exceptional capability to ingest and breakdown dense, sophisticated text data quickly.
• Nuanced Decision Making: High comfort level operating within ambiguous contexts, using partial data or conflicting guidelines to form sound, objective conclusions.
• Critical Evaluation: Innate analytical skepticism needed to separate biased or unsafe outputs from objective text.
  
  

GenAI Behavioral & Cognitive Readiness

• Cognitive Resilience: Proven ability to maintain speed and precision when processing high-volume data streams under demanding cognitive load.
• Learning Agility: Proactive curiosity regarding AI technologies, with a natural inclination to adapt swiftly to changing model updates, rules, and ecosystem dynamics.
• Ethical Judgment & Bias Awareness: Strict commitment to personal integrity, maintaining high emotional intelligence to detect structural or societal bias within AI behavior.
  
  

Technical Environment

Additional Information

Career Level

Entry to Mid

Work Location

Quezon City

Work Setup

Fully onsite

Job Type

Project based

APPLY NOW