Hire this AI Trainer
Sign in or create an account to invite AI Trainers to your job.
AWS SageMaker
Clickworker
CrowdSource
Labelbox
Scale AINo subject matter listed
Audio
Computer Code Programming
DocumentI have extensive experience in AI training data and model evaluation, having contributed to the assessment of multiple large-scale models for leading organizations including OpenAI, Google, and Meta AI. My primary expertise lies in evaluating mathematical reasoning and computer programming outputs, where precision, logical consistency, and adherence to specifications are critical. In addition, I have worked on projects involving text, function calling, audio, and visual data, giving me a well-rounded perspective on multimodal AI evaluation. Beyond commercial model evaluation, I have collaborated with university research teams on developing AI models for text simplification in both English and Estonian. This work required careful annotation, linguistic sensitivity, and alignment with research objectives, further strengthening my ability to produce high-quality, reliable training data across diverse domains and use cases.
The models were tasked with solving mathematical problems from an uploaded image. The main goal was to evaluate the correctness of both the solutions and the capturing of information from the image.
Goal of the project was to evaluate AI generated video outputs according to the given prompt. Main focus was that everything from the prompt needed to be included and for the video to not include any hallucinations.
The goal of the project was to ask the model to create a website with specific key components. Then three different models were evaluated side by side for accuracy and professionalism.
The goal of the project was to write a mathematical problem complicated enough to make the model fail to give a correct solution. Then rate the model across fields like text and localization and finally correct it's mistakes.
The main goal was to talk with an AI assistant and ensure that the model fails to assist the user on certain requests. Then rate the model for fields like speed, tone, information correctness etc.
Bachelor, Computer Science
Gymnasium Degree, Real Sciences
Research Programmer
CEO and Lead Developer
