AI Developer | Full Stack Developer | ISV Success | DevOps Specialist

leetcode

youtube

devto

medium

github

Zack E. • Peerlist

Cell seeding efficiency measurement  using customized Cellpose SAM (cellpose4) models. The goal is to classify cells into four categories and calculate cell properties (number, average brightness, and average area) for each frame

Cellpose Sam Model Cell Seeding

This project SNAPSHOT introduces a GitHub Copilot Extension integrated with Cloud Groq APIs , utilizing powerful AI models based on @copilot-extensions github models extension

Copilot Extension Groq

Create custom action scripts on Github in .YAML format in different scenario development modes: deploy, package checkout, pull and install,unit test,...

DevOps Developer - Github Actions

Digital marketing platform for influencers that can significantly optimize campaigns, streamline processes, and drive better results

Digital Marketing

Hotel Bookings Web App Manage based on both versions ,{ Javascript ,CSS ,HTML ,Php } version and { React , Blade , Php  } version.

Hotel Bookings

Set up and operated LLM workspaces, focusing on Ingesting Data Efficiently managed the ingestion of large datasets into the LLM workspace, ensuring data integrity and relevance ,Configured the LLM environment, including model selection, parameter tuning, and integration with existing systems, ensuring seamless operation ,Executed LLM models for various applications, monitoring performance and optimizing for efficiency and accuracy, Continuously refined the LLM setup to enhance performance, addressing issues related to model accuracy and computational resources. Maintained detailed documentation of LLM workspace configurations and generated reports on model performance and improvements.

LLM Workspace

Successfully integrated two applications, connecting a payment application with PayPal through the PayPal API via Sandbox mode, improving transaction efficiency and streamlining payment processing.

PayPal Developer

with a focus on multi-turn conversations involving prompts and responses. Instructions were followed with high accuracy, without any issues. Code brevity and relevance were addressed, with minor issues occasionally encountered due to complex code structures. Safety standards were maintained without any major issues.Annotations were presented in a clear and organized format, without any issues. Overall high quality was achieved, with minor issues occasionally arising due to the need for iterative improvement. Additionally, LLM model responses were parsed using JSON schemas, with a focus on input, thought chain, and output arrays.

Prompt Engineering

<p>Advanced vision transformers and segmentation models (DINO, ViT, SAM, and YOLO+) to build a robust pipeline for identifying and isolating trees from images . With refined post-processing steps and enhanced mask filtering, high precision segmentation , improving overall performance through metrics such as IoU, precision, recall, and AP.</p>