EXPLORE

0.0

(0 reviews)

Serhii H.

@serhiih25

0.0

0

0.0

0.0

0%

Solution Solver

$30 USD / Hour

・

Ukraine (12:33 PM)

・

Joined on November 12, 2025

$30 USD / Hour

・

As the Co-Founder and Chief Engineer of Gondar Tech, I lead a world-class team specializing in generative AI platform development. With years of experience in AI model development, MLOps, and full-stack integration, I have successfully designed and deployed scalable AI systems for enterprises and innovators globally. I help businesses and entrepreneurs harness the full power of Generative AI — from text, image, and video generation to custom AI model training and deployment — providing solutions that are scalable, secure, and cutting-edge. What I offer: - Development and deployment of custom generative AI platform.End-to-end AI integration for web, mobile, and cloud application. - MLOps expertise for model versioning, scaling, and optimization. - Consultation on AI strategy and AI product innovation. - Team leadership in complex AI projects, ensuring high-quality deliverables. I am passionate about building the future of AI and bringing innovative, practical AI solutions to life for companies around the world.

Verifications

See all PHP Developers in Ukraine

Portfolio

Portfolio

Developed a comprehensive, production-ready frontend for a multi-tenant SaaS dental clinic management system using Next.js 15 and TypeScript. This enterprise-grade application features four distinct role-based dashboards (Super Admin, Clinic Admin, Dentist, and Receptionist) with complete patient management, appointment scheduling, and dental charting capabilities.

Key achievements include implementing JWT-based authentication with role-based access control, creating an interactive 3D dental charting system with AI-powered diagnosis assistance, building a responsive UI with 50+ reusable components using shadcn/ui and Tailwind CSS, and integrating comprehensive features like invoice management, inventory tracking, SMS notifications, and financial reporting.

The application leverages Next.js App Router for optimal performance, implements server and client components strategically, and follows modern React patterns with full TypeScript coverage. Built with scalability in mind, it suppor

SaaS Dental Clinic Management System

Developed a comprehensive, production-ready frontend for a multi-tenant SaaS dental clinic management system using Next.js 15 and TypeScript. This enterprise-grade application features four distinct role-based dashboards (Super Admin, Clinic Admin, Dentist, and Receptionist) with complete patient management, appointment scheduling, and dental charting capabilities.

Key achievements include implementing JWT-based authentication with role-based access control, creating an interactive 3D dental charting system with AI-powered diagnosis assistance, building a responsive UI with 50+ reusable components using shadcn/ui and Tailwind CSS, and integrating comprehensive features like invoice management, inventory tracking, SMS notifications, and financial reporting.

The application leverages Next.js App Router for optimal performance, implements server and client components strategically, and follows modern React patterns with full TypeScript coverage. Built with scalability in mind, it suppor

SaaS Dental Clinic Management System

Developed a comprehensive, production-ready frontend for a multi-tenant SaaS dental clinic management system using Next.js 15 and TypeScript. This enterprise-grade application features four distinct role-based dashboards (Super Admin, Clinic Admin, Dentist, and Receptionist) with complete patient management, appointment scheduling, and dental charting capabilities.

Key achievements include implementing JWT-based authentication with role-based access control, creating an interactive 3D dental charting system with AI-powered diagnosis assistance, building a responsive UI with 50+ reusable components using shadcn/ui and Tailwind CSS, and integrating comprehensive features like invoice management, inventory tracking, SMS notifications, and financial reporting.

The application leverages Next.js App Router for optimal performance, implements server and client components strategically, and follows modern React patterns with full TypeScript coverage. Built with scalability in mind, it suppor

SaaS Dental Clinic Management System

Developed a comprehensive, production-ready frontend for a multi-tenant SaaS dental clinic management system using Next.js 15 and TypeScript. This enterprise-grade application features four distinct role-based dashboards (Super Admin, Clinic Admin, Dentist, and Receptionist) with complete patient management, appointment scheduling, and dental charting capabilities.

Key achievements include implementing JWT-based authentication with role-based access control, creating an interactive 3D dental charting system with AI-powered diagnosis assistance, building a responsive UI with 50+ reusable components using shadcn/ui and Tailwind CSS, and integrating comprehensive features like invoice management, inventory tracking, SMS notifications, and financial reporting.

The application leverages Next.js App Router for optimal performance, implements server and client components strategically, and follows modern React patterns with full TypeScript coverage. Built with scalability in mind, it suppor

SaaS Dental Clinic Management System

Developed simple generative AI service demo and integrated to my past portfolio site.
It is shared on my github as public repository.
Used base diffusion models and pipelines and it was not live now.

Portfolio Site V1 (gen ai part)

Developed simple generative AI service demo and integrated to my past portfolio site.
It is shared on my github as public repository.
Used base diffusion models and pipelines and it was not live now.

Portfolio Site V1 (gen ai part)

Developed simple generative AI service demo and integrated to my past portfolio site.
It is shared on my github as public repository.
Used base diffusion models and pipelines and it was not live now.

Portfolio Site V1 (gen ai part)

Developed simple generative AI service demo and integrated to my past portfolio site.
It is shared on my github as public repository.
Used base diffusion models and pipelines and it was not live now.

Portfolio Site V1 (gen ai part)

Developed simple generative AI service demo and integrated to my past portfolio site.
It is shared on my github as public repository.
Used base diffusion models and pipelines and it was not live now.

Portfolio Site V1 (gen ai part)

Developed simple generative AI service demo and integrated to my past portfolio site.
It is shared on my github as public repository.
Used base diffusion models and pipelines and it was not live now.

Portfolio Site V1 (gen ai part)

Implemented chatbot with RAG pipeline using Open-Source (Qwen-2.5) and Closed-Source (OpenAI GPT 4o) models.
Shared full architecture of this app in the uploaded images.

Portfolio Site V1 (chatbot part)

Implemented chatbot with RAG pipeline using Open-Source (Qwen-2.5) and Closed-Source (OpenAI GPT 4o) models.
Shared full architecture of this app in the uploaded images.

Portfolio Site V1 (chatbot part)

Implemented chatbot with RAG pipeline using Open-Source (Qwen-2.5) and Closed-Source (OpenAI GPT 4o) models.
Shared full architecture of this app in the uploaded images.

Portfolio Site V1 (chatbot part)

Implemented chatbot with RAG pipeline using Open-Source (Qwen-2.5) and Closed-Source (OpenAI GPT 4o) models.
Shared full architecture of this app in the uploaded images.

Portfolio Site V1 (chatbot part)

Implemented chatbot with RAG pipeline using Open-Source (Qwen-2.5) and Closed-Source (OpenAI GPT 4o) models.
Shared full architecture of this app in the uploaded images.

Portfolio Site V1 (chatbot part)

Implemented chatbot with RAG pipeline using Open-Source (Qwen-2.5) and Closed-Source (OpenAI GPT 4o) models.
Shared full architecture of this app in the uploaded images.

Portfolio Site V1 (chatbot part)

Configured Runpod Serverless endpoint for VTON (virtual try-on).
Should used open-source comfyui workflow and runpod handler docker image build method.

Virtual Try-on Serverless Endpoint

Configured Runpod Serverless endpoint for VTON (virtual try-on).
Should used open-source comfyui workflow and runpod handler docker image build method.

Virtual Try-on Serverless Endpoint

Configured Runpod Serverless endpoint for VTON (virtual try-on).
Should used open-source comfyui workflow and runpod handler docker image build method.

Virtual Try-on Serverless Endpoint

To develop this app, I took part in AI part and worked for data cleaning, modeling, model tuning, serverless computing service deplyment.
Used flux model and other specific LoRA training method and used Runpod Serverless as Cloud Solution.
My most valuable contribution was that I reduced inference time of pipeline from 24~32s to 16~19s using model caching algorithm.

Voila AI Artist - AI Part Dev

To develop this app, I took part in AI part and worked for data cleaning, modeling, model tuning, serverless computing service deplyment.
Used flux model and other specific LoRA training method and used Runpod Serverless as Cloud Solution.
My most valuable contribution was that I reduced inference time of pipeline from 24~32s to 16~19s using model caching algorithm.

Voila AI Artist - AI Part Dev

To develop this app, I took part in AI part and worked for data cleaning, modeling, model tuning, serverless computing service deplyment.
Used flux model and other specific LoRA training method and used Runpod Serverless as Cloud Solution.
My most valuable contribution was that I reduced inference time of pipeline from 24~32s to 16~19s using model caching algorithm.

Voila AI Artist - AI Part Dev

To develop this app, I took part in AI part and worked for data cleaning, modeling, model tuning, serverless computing service deplyment.
Used flux model and other specific LoRA training method and used Runpod Serverless as Cloud Solution.
My most valuable contribution was that I reduced inference time of pipeline from 24~32s to 16~19s using model caching algorithm.

Voila AI Artist - AI Part Dev

To develop this app, I took part in AI part and worked for data cleaning, modeling, model tuning, serverless computing service deplyment.
Used flux model and other specific LoRA training method and used Runpod Serverless as Cloud Solution.
My most valuable contribution was that I reduced inference time of pipeline from 24~32s to 16~19s using model caching algorithm.

Voila AI Artist - AI Part Dev

To develop this app, I took part in AI part and worked for data cleaning, modeling, model tuning, serverless computing service deplyment.
Used flux model and other specific LoRA training method and used Runpod Serverless as Cloud Solution.
My most valuable contribution was that I reduced inference time of pipeline from 24~32s to 16~19s using model caching algorithm.

Voila AI Artist - AI Part Dev

To develop this app, I took part in AI part and worked for data cleaning, modeling, model tuning, serverless computing service deplyment.
Used flux model and other specific LoRA training method and used Runpod Serverless as Cloud Solution.
My most valuable contribution was that I reduced inference time of pipeline from 24~32s to 16~19s using model caching algorithm.

Voila AI Artist - AI Part Dev

To develop this app, I took part in AI part and worked for data cleaning, modeling, model tuning, serverless computing service deplyment.
Used flux model and other specific LoRA training method and used Runpod Serverless as Cloud Solution.
My most valuable contribution was that I reduced inference time of pipeline from 24~32s to 16~19s using model caching algorithm.

Voila AI Artist - AI Part Dev

To develop this app, I took part in AI part and worked for data cleaning, modeling, model tuning, serverless computing service deplyment.
Used flux model and other specific LoRA training method and used Runpod Serverless as Cloud Solution.
My most valuable contribution was that I reduced inference time of pipeline from 24~32s to 16~19s using model caching algorithm.

Voila AI Artist - AI Part Dev

Developed a comprehensive, production-ready frontend for a multi-tenant SaaS dental clinic management system using Next.js 15 and TypeScript. This enterprise-grade application features four distinct role-based dashboards (Super Admin, Clinic Admin, Dentist, and Receptionist) with complete patient management, appointment scheduling, and dental charting capabilities.

Key achievements include implementing JWT-based authentication with role-based access control, creating an interactive 3D dental charting system with AI-powered diagnosis assistance, building a responsive UI with 50+ reusable components using shadcn/ui and Tailwind CSS, and integrating comprehensive features like invoice management, inventory tracking, SMS notifications, and financial reporting.

The application leverages Next.js App Router for optimal performance, implements server and client components strategically, and follows modern React patterns with full TypeScript coverage. Built with scalability in mind, it suppor

SaaS Dental Clinic Management System

Developed a comprehensive, production-ready frontend for a multi-tenant SaaS dental clinic management system using Next.js 15 and TypeScript. This enterprise-grade application features four distinct role-based dashboards (Super Admin, Clinic Admin, Dentist, and Receptionist) with complete patient management, appointment scheduling, and dental charting capabilities.

Key achievements include implementing JWT-based authentication with role-based access control, creating an interactive 3D dental charting system with AI-powered diagnosis assistance, building a responsive UI with 50+ reusable components using shadcn/ui and Tailwind CSS, and integrating comprehensive features like invoice management, inventory tracking, SMS notifications, and financial reporting.

The application leverages Next.js App Router for optimal performance, implements server and client components strategically, and follows modern React patterns with full TypeScript coverage. Built with scalability in mind, it suppor

SaaS Dental Clinic Management System

Developed a comprehensive, production-ready frontend for a multi-tenant SaaS dental clinic management system using Next.js 15 and TypeScript. This enterprise-grade application features four distinct role-based dashboards (Super Admin, Clinic Admin, Dentist, and Receptionist) with complete patient management, appointment scheduling, and dental charting capabilities.

Key achievements include implementing JWT-based authentication with role-based access control, creating an interactive 3D dental charting system with AI-powered diagnosis assistance, building a responsive UI with 50+ reusable components using shadcn/ui and Tailwind CSS, and integrating comprehensive features like invoice management, inventory tracking, SMS notifications, and financial reporting.

The application leverages Next.js App Router for optimal performance, implements server and client components strategically, and follows modern React patterns with full TypeScript coverage. Built with scalability in mind, it suppor

SaaS Dental Clinic Management System

Developed a comprehensive, production-ready frontend for a multi-tenant SaaS dental clinic management system using Next.js 15 and TypeScript. This enterprise-grade application features four distinct role-based dashboards (Super Admin, Clinic Admin, Dentist, and Receptionist) with complete patient management, appointment scheduling, and dental charting capabilities.

Key achievements include implementing JWT-based authentication with role-based access control, creating an interactive 3D dental charting system with AI-powered diagnosis assistance, building a responsive UI with 50+ reusable components using shadcn/ui and Tailwind CSS, and integrating comprehensive features like invoice management, inventory tracking, SMS notifications, and financial reporting.

The application leverages Next.js App Router for optimal performance, implements server and client components strategically, and follows modern React patterns with full TypeScript coverage. Built with scalability in mind, it suppor

SaaS Dental Clinic Management System

Developed simple generative AI service demo and integrated to my past portfolio site.
It is shared on my github as public repository.
Used base diffusion models and pipelines and it was not live now.

Portfolio Site V1 (gen ai part)

Developed simple generative AI service demo and integrated to my past portfolio site.
It is shared on my github as public repository.
Used base diffusion models and pipelines and it was not live now.

Portfolio Site V1 (gen ai part)

Developed simple generative AI service demo and integrated to my past portfolio site.
It is shared on my github as public repository.
Used base diffusion models and pipelines and it was not live now.

Portfolio Site V1 (gen ai part)

Developed simple generative AI service demo and integrated to my past portfolio site.
It is shared on my github as public repository.
Used base diffusion models and pipelines and it was not live now.

Portfolio Site V1 (gen ai part)

Developed simple generative AI service demo and integrated to my past portfolio site.
It is shared on my github as public repository.
Used base diffusion models and pipelines and it was not live now.

Portfolio Site V1 (gen ai part)

Developed simple generative AI service demo and integrated to my past portfolio site.
It is shared on my github as public repository.
Used base diffusion models and pipelines and it was not live now.

Portfolio Site V1 (gen ai part)

Implemented chatbot with RAG pipeline using Open-Source (Qwen-2.5) and Closed-Source (OpenAI GPT 4o) models.
Shared full architecture of this app in the uploaded images.

Portfolio Site V1 (chatbot part)

Implemented chatbot with RAG pipeline using Open-Source (Qwen-2.5) and Closed-Source (OpenAI GPT 4o) models.
Shared full architecture of this app in the uploaded images.

Portfolio Site V1 (chatbot part)

Implemented chatbot with RAG pipeline using Open-Source (Qwen-2.5) and Closed-Source (OpenAI GPT 4o) models.
Shared full architecture of this app in the uploaded images.

Portfolio Site V1 (chatbot part)

Implemented chatbot with RAG pipeline using Open-Source (Qwen-2.5) and Closed-Source (OpenAI GPT 4o) models.
Shared full architecture of this app in the uploaded images.

Portfolio Site V1 (chatbot part)

Implemented chatbot with RAG pipeline using Open-Source (Qwen-2.5) and Closed-Source (OpenAI GPT 4o) models.
Shared full architecture of this app in the uploaded images.

Portfolio Site V1 (chatbot part)

Implemented chatbot with RAG pipeline using Open-Source (Qwen-2.5) and Closed-Source (OpenAI GPT 4o) models.
Shared full architecture of this app in the uploaded images.

Portfolio Site V1 (chatbot part)

Configured Runpod Serverless endpoint for VTON (virtual try-on).
Should used open-source comfyui workflow and runpod handler docker image build method.

Virtual Try-on Serverless Endpoint

Configured Runpod Serverless endpoint for VTON (virtual try-on).
Should used open-source comfyui workflow and runpod handler docker image build method.

Virtual Try-on Serverless Endpoint

Configured Runpod Serverless endpoint for VTON (virtual try-on).
Should used open-source comfyui workflow and runpod handler docker image build method.

Virtual Try-on Serverless Endpoint

To develop this app, I took part in AI part and worked for data cleaning, modeling, model tuning, serverless computing service deplyment.
Used flux model and other specific LoRA training method and used Runpod Serverless as Cloud Solution.
My most valuable contribution was that I reduced inference time of pipeline from 24~32s to 16~19s using model caching algorithm.

Voila AI Artist - AI Part Dev

To develop this app, I took part in AI part and worked for data cleaning, modeling, model tuning, serverless computing service deplyment.
Used flux model and other specific LoRA training method and used Runpod Serverless as Cloud Solution.
My most valuable contribution was that I reduced inference time of pipeline from 24~32s to 16~19s using model caching algorithm.

Voila AI Artist - AI Part Dev

To develop this app, I took part in AI part and worked for data cleaning, modeling, model tuning, serverless computing service deplyment.
Used flux model and other specific LoRA training method and used Runpod Serverless as Cloud Solution.
My most valuable contribution was that I reduced inference time of pipeline from 24~32s to 16~19s using model caching algorithm.

Voila AI Artist - AI Part Dev

To develop this app, I took part in AI part and worked for data cleaning, modeling, model tuning, serverless computing service deplyment.
Used flux model and other specific LoRA training method and used Runpod Serverless as Cloud Solution.
My most valuable contribution was that I reduced inference time of pipeline from 24~32s to 16~19s using model caching algorithm.

Voila AI Artist - AI Part Dev

To develop this app, I took part in AI part and worked for data cleaning, modeling, model tuning, serverless computing service deplyment.
Used flux model and other specific LoRA training method and used Runpod Serverless as Cloud Solution.
My most valuable contribution was that I reduced inference time of pipeline from 24~32s to 16~19s using model caching algorithm.

Voila AI Artist - AI Part Dev

To develop this app, I took part in AI part and worked for data cleaning, modeling, model tuning, serverless computing service deplyment.
Used flux model and other specific LoRA training method and used Runpod Serverless as Cloud Solution.
My most valuable contribution was that I reduced inference time of pipeline from 24~32s to 16~19s using model caching algorithm.

Voila AI Artist - AI Part Dev

To develop this app, I took part in AI part and worked for data cleaning, modeling, model tuning, serverless computing service deplyment.
Used flux model and other specific LoRA training method and used Runpod Serverless as Cloud Solution.
My most valuable contribution was that I reduced inference time of pipeline from 24~32s to 16~19s using model caching algorithm.

Voila AI Artist - AI Part Dev

To develop this app, I took part in AI part and worked for data cleaning, modeling, model tuning, serverless computing service deplyment.
Used flux model and other specific LoRA training method and used Runpod Serverless as Cloud Solution.
My most valuable contribution was that I reduced inference time of pipeline from 24~32s to 16~19s using model caching algorithm.

Voila AI Artist - AI Part Dev

To develop this app, I took part in AI part and worked for data cleaning, modeling, model tuning, serverless computing service deplyment.
Used flux model and other specific LoRA training method and used Runpod Serverless as Cloud Solution.
My most valuable contribution was that I reduced inference time of pipeline from 24~32s to 16~19s using model caching algorithm.

Voila AI Artist - AI Part Dev

Reviews

Reviews

No reviews to see here!

Co-Founder

Nov, 2025 - Present

•

4 months, 1 day

Gondar Tech

Nov, 2025 - Present

•

4 months, 1 day

I'm working as Co-Founder and Chief Engineer.

Berezan, Ukraine

Nov, 2025 - Present

•

4 months, 1 day

AI Software Engineer

Dec, 2023 - Oct, 2025

•

1 year, 10 months

DataArt

Dec, 2023 - Oct, 2025

•

1 year, 10 months

Worked as Software Engineer

Berezan, Ukraine

Dec, 2023 - Oct, 2025

•

1 year, 10 months

Software Developer

May, 2021 - Sep, 2023

•

2 years, 3 months

DataArt

May, 2021 - Sep, 2023

•

2 years, 3 months

Worked as AI Software Engineer

Berezan, Ukraine

May, 2021 - Sep, 2023

•

2 years, 3 months

Odessa National 'I.I. Mecnikov' University

2017 - 2021

•

4 years

Bachelor

Ukraine

2017 - 2021

•

4 years

P

Preferred Freelancer Program SLA