Looks like you have JavaScript disabled. For the full Twine experience, you will need to re-enable it.

I have extensive in machine learning and deploying AI systems. Specifically, I worked on computer vision, natural language processing, graph neural networks, and machine learning optimization techniques. For computer vision, I worked on improving the model's ability to segment and classify an image by 1) re-designing the neural network architecture by retaining learned information on earlier layers and aggregating it to information learned by more advanced layers to be able to segment the image more efficiently (think about multi-task learning by having a parallel neural network architecture with one architecture for segmenting and the main architecture for classifying). 2) By reviewing datasets and proof-checking that the images are of high resolution, and labeled correctly. For optimizing machine learning models, I explored and applied optimization methods to reduce the machine learning model's size and increase inference speed while maintaining the overall weighted accuracy of the model. This was applied to computer vision and graph neural network models. I focused on applying pruning, quantization, and early exiting methods. For natural language processing, I worked on 1) evaluating an LLM prompt and response, and checking if the LLM is responding correctly or is hallucinating some information. 2) I created voice datasets for AI models to train the AI to be more conversational and speak more naturally like humans. 3) Re-train and fine-tune the LLM to learn from its mistakes and improve.…I have extensive in machine learning and deploying AI systems. Specifically, I worked on computer vision, natural language processing, graph neural networks, and machine learning optimization techniques. For computer vision, I worked on improving the model's ability to segment and classify an image by 1) re-designing the neural network architecture by retaining learned information on earlier layers and aggregating it to information learned by more advanced layers to be able to segment the image more efficiently (think about multi-task learning by having a parallel neural network architecture with one architecture for segmenting and the main architecture for classifying). 2) By reviewing datasets and proof-checking that the images are of high resolution, and labeled correctly. For optimizing machine learning models, I explored and applied optimization methods to reduce the machine learning model's size and increase inference speed while maintaining the overall weighted accuracy of the model. This was applied to computer vision and graph neural network models. I focused on applying pruning, quantization, and early exiting methods. For natural language processing, I worked on 1) evaluating an LLM prompt and response, and checking if the LLM is responding correctly or is hallucinating some information. 2) I created voice datasets for AI models to train the AI to be more conversational and speak more naturally like humans. 3) Re-train and fine-tune the LLM to learn from its mistakes and improve.

AlyElsayed

Data Scientist, AI Engineer, Cloud Developer, +1





I have extensive in machine learning and deploying AI systems. Specifically, I worked on computer vision, natural language processing, graph neural networks, and machine learning optimization techniques. For computer vision, I worked on improving the model's ability to segment and classify an image by 1) re-designing the neural network architecture by retaining learned information on earlier layers and aggregating it to information learned by more advanced layers to be able to segment the image more efficiently (think about multi-task learning by having a parallel neural network architecture with one architecture for segmenting and the main architecture for classifying). 2) By reviewing datasets and proof-checking that the images are of high resolution, and labeled correctly. For optimizing machine learning models, I explored and applied optimization methods to reduce the machine learning model's size and increase inference speed while maintaining the overall weighted accuracy of the model. This was applied to computer vision and graph neural network models. I focused on applying pruning, quantization, and early exiting methods. For natural language processing, I worked on 1) evaluating an LLM prompt and response, and checking if the LLM is responding correctly or is hallucinating some information. 2) I created voice datasets for AI models to train the AI to be more conversational and speak more naturally like humans. 3) Re-train and fine-tune the LLM to learn from its mistakes and improve.…I have extensive in machine learning and deploying AI systems. Specifically, I worked on computer vision, natural language processing, graph neural networks, and machine learning optimization techniques. For computer vision, I worked on improving the model's ability to segment and classify an image by 1) re-designing the neural network architecture by retaining learned information on earlier layers and aggregating it to information learned by more advanced layers to be able to segment the image more efficiently (think about multi-task learning by having a parallel neural network architecture with one architecture for segmenting and the main architecture for classifying). 2) By reviewing datasets and proof-checking that the images are of high resolution, and labeled correctly. For optimizing machine learning models, I explored and applied optimization methods to reduce the machine learning model's size and increase inference speed while maintaining the overall weighted accuracy of the model. This was applied to computer vision and graph neural network models. I focused on applying pruning, quantization, and early exiting methods. For natural language processing, I worked on 1) evaluating an LLM prompt and response, and checking if the LLM is responding correctly or is hallucinating some information. 2) I created voice datasets for AI models to train the AI to be more conversational and speak more naturally like humans. 3) Re-train and fine-tune the LLM to learn from its mistakes and improve.

Available to hire

I have extensive in machine learning and deploying AI systems. Specifically, I worked on computer vision, natural language processing, graph neural networks, and machine learning optimization techniques.

For computer vision, I worked on improving the model’s ability to segment and classify an image by 1) re-designing the neural network architecture by retaining learned information on earlier layers and aggregating it to information learned by more advanced layers to be able to segment the image more efficiently (think about multi-task learning by having a parallel neural network architecture with one architecture for segmenting and the main architecture for classifying). 2) By reviewing datasets and proof-checking that the images are of high resolution, and labeled correctly.

For optimizing machine learning models, I explored and applied optimization methods to reduce the machine learning model’s size and increase inference speed while maintaining the overall weighted accuracy of the model. This was applied to computer vision and graph neural network models. I focused on applying pruning, quantization, and early exiting methods.

For natural language processing, I worked on 1) evaluating an LLM prompt and response, and checking if the LLM is responding correctly or is hallucinating some information. 2) I created voice datasets for AI models to train the AI to be more conversational and speak more naturally like humans. 3) Re-train and fine-tune the LLM to learn from its mistakes and improve.