Hire this AI Trainer
Sign in or create an account to invite AI Trainers to your job.
Text
DocumentAI Telco Troubleshooting Challenge (LLM Fine-tuning). Brings 7+ years of professional experience across complex professional workflows, research, and quality-focused execution. Core strengths include Other. Education includes Certificate, Flatiron School (2025) and Bachelor of Science, Dedan Kimathi University of Technology (2019). AI-training focus includes data types such as Text and labeling workflows including Fine-tuning.
Review multi-turn conversations, evaluate pre-provided rubric criteria against three LLM chatbot model responses, edit rubrics when necessary, and rate each rubric on quality dimensions.
Fine-tuned the RoBERTa transformer to categorize sentiment in user posts about Google and Apple products. Responsible for preparing labeled training data, training the model, and evaluating the effectiveness. Achieved a high macro F1-score demonstrating the quality of the labeling and fine-tuning process. • Curated and labeled sentiment data from Twitter • Implemented data preprocessing and model evaluation steps • Utilized Python and deep learning software for experiments • Successfully increased accuracy of sentiment classification
Fine-tuned SciBERT and MathBERTa transformer models to identify and classify specific pedagogical errors in student responses to K-12 math problems. The project focused on error analysis for improving educational AI systems. Extensive data labeling and preparation were performed to support effective model fine-tuning. • Labeled and pre-processed K-12 student response data for model training • Applied fine-tuning techniques to transformer architectures • Used Python and deep learning libraries for all modeling tasks • Enhanced the ability to detect nuanced student misunderstandings
Fine-tuned the Qwen-2.5 1.5-Instruct large language model using QLoRA (PEFT) for automated root-cause analysis in telecommunications troubleshooting. The project aimed to enhance the model's ability to classify suboptimal network signals with increased accuracy and efficiency. The process involved data preparation, model training, and evaluation using advanced Python and deep learning frameworks. • Automating root-cause analysis tasks in network signal troubleshooting • Data set curation and preparation for fine-tuning a language model • Employed QLoRA, Python, and modern deep learning libraries • Measured improvements in classification accuracy and memory efficiency
Certificate, Data Science
Bachelor of Science, Electrical and Electronics Engineering
Maintenance Department Intern
Electrical Technician Apprentice
