Hire this AI Trainer
Sign in or create an account to invite AI Trainers to your job.
Appen
Clickworker
Labelbox
OneForma
Remotasks
Scale AI
SuperAnnotate
TolokaNo subject matter listed
Computer Code Programming
Image
TextFor the past two years I’ve specialised in human-in-the-loop training of large-language models, supplying >6 000 high-quality feedback samples that power code assistants and chatbots. My core work spans RLHF — ranking and critiquing model-generated code, diagnosing compiler/runtime errors, and crafting gold-standard fixes — using pipelines on Scale AI, Toloka and Labelbox. These datasets feed reward models that raise functional accuracy, reduce bias, and harden models against unsafe outputs. In parallel, I lead multilingual prompt QA (English ↔ Arabic) and machine-translation post-editing. My safety focus includes extensive red-teaming and jailbreak testing—simulating adversarial prompts that expose bias or policy leaks before deployment.
Labeled 500+ tool-use chains (arguments, returns, error states) that drive fine-tuning and regression tests for API-enabled chat agents.
Post-edited 100 k+ words (EN↔AR) in MTPE workflows and ran content-safety sweeps, cutting critical error rate below 2 %
Audited Arabic/English prompts for tone and cultural fit; produced the style guide now used across OneForma’s ISAAC translation projects
Ranked and critiqued 900+ code generations, authored reference solutions and unit tests, and annotated JSON function-call schemas
Bachelor's Degree, Computers And Informatics
Principal AI Architect (Consultant)
Senior WordPress & Automation Consultant
