endorsedCompleted

Winbets.ai MVP — AI Chatbot & Full-Stack Development

Services Covered:

Custom Software Development • AI and Machine Learning • Website Development

Industry

Sports & Entertainment

Duration

8 months

Budget

Confidential

Start Date

1 January 2024

Team Size

6-10 Employees

Engagement Model

Offshore

Project Capability Score (PCS)

81/100

Strong

Project Capability Score (PCS) estimates how capable Viso is of successfully delivering a project like this, based on its past experience, track record, and ability to handle similar work. Learn more about PCS

Client

Winbets.ai

San Diego, United States of America

Project Summary

Viso served as the primary development partner for the Winbets.ai MVP, delivering full-stack web development, AI chatbot integration, system architecture planning, and QA/testing. The vendor built core platform features including real-time odds integration, user authentication, and a conversational AI assistant. The team delivered the MVP on a tight timeline with solid test coverage and deployment pipelines, and was noted for responsiveness and technical acumen.

Key Challenges

Full-scale MVP launch within 7 months from hire to launch
Build iOS and Android apps from concept to launch
Prepare and conduct AI training and integration for predictions

Project Deliverables

Backend API using Next.js with Prisma and tRPC
Frontend UI built with Next.js, TailwindCSS and ShadCN components
Deployment pipelines configured on Vercel with GitHub Actions
OAuth2 authentication implementation (Google/Apple via Next-auth)
Database schemas for users, odds, matches, and chats (PostgreSQL)
Chat UI and Clutch AI assistant integration with OpenAI GPT models
Gradient boosting prediction model (Python / Scikit-learn)
Real-time odds integration via WebSocket/Webhook
Usage logging and chat history storage for AI fine-tuning
Automated unit and end-to-end tests (70%+ UI coverage)
Performance optimization and launch preparation

Project Solution

VISO was contracted as the primary development partner to build the Winbets.ai MVP. They executed end-to-end full-stack development using Next.js and TypeScript, implemented OAuth2 authentication (Google/Apple), modeled and migrated PostgreSQL schemas, and set up Vercel deployment pipelines with GitHub Actions. The team integrated OpenAI GPT models for a conversational assistant (Clutch), developed a Gradient Boosting classifier in Python (Scikit-learn) for predictions, connected prediction logic to the chat via function-calling, integrated live odds feeds, implemented usage logging for future AI training, and built automated unit and end-to-end tests and performance optimizations ahead of launch.

Project Outcome

VISO delivered the full Winbets.ai MVP within a tight 7–9 month timeline
Built core platform including real-time odds integration, user authentication, and a working AI assistant
Implemented deployment pipelines and achieved solid test coverage and backend infrastructure for logging and AI training
Client reported the team was responsive, technically sharp, and consistently met milestones

Platforms

Mobile
Web
Cloud

Tech Stack

Next.js
TypeScript
Tailwind CSS
PostgreSQL
Python
Scikit-Learn
Vercel
GitHub Actions

Client Endorsement

Overall Review Rating

Timeliness

Cost Rating

Willing to Refer

Quality of Deliverables

“Viso successfully delivered the MVP within a tight timeline. The team was responsive, technically sharp, and consistently hit key milestones. Ultimately, Visos' impressive work, combined with their solid test coverage and backend infrastructure, made them a valuable partner.”
Mark Krause
CEO/Founder
Date: 05/06/2025