AI Engineer
ITC International
Build AI projects, chatbots and automation, red-team and evaluate AI models, and create training data.
Key outcomes & impact
- Build LLM-powered applications, chatbots and Python automation workflows end to end, from prompt and pipeline design to integration and testing.
- Create training and evaluation datasets for AI models, writing challenging multilingual prompts across formats, languages and Arabic dialects to target known and new failure modes.
- Design adversarial test scenarios (AI red teaming) that probe LLM reliability, factual accuracy, hallucination, bias and unsafe behavior.
- Evaluate model outputs against a structured rubric (accuracy, relevance, completeness, neutrality, safety, source validation) and trace the root cause of each failure with supporting evidence.
- Deliver reproducible defect reports (prompt, response, weakness, recommended correction) that drive model improvements and technical investigation.