Home / Case Studies

Applied AI & Product Case Studies

Engineering breakdowns of independent products across cost modeling, local CPU voice inference, and browser-based AI workflows.

Case Study 01

AgentStack Calc (agentstackcalc.com)

Role: Creator & Lead Engineer · May 2026
Next.js App Router Markdown Compiler Dynamic GSC Indexing

The Challenge

Estimating operating costs for speech AI, language-model usage, call routing, and automation is difficult because pricing is spread across multiple vendors and billing units.

AgentStack Calc brings those inputs into one interactive simulator so teams can compare managed APIs and self-hosted workflow options before committing to a stack.

Engineering Solution

  • Static Generation (SSG) & Local Execution: Programmed simulator states to calculate client-side in memory, keeping comparisons responsive without requiring a database query for each change.
  • Decoupled Markdown Parser: Configured a dynamic filesystem reader inside Next.js using gray-matter and marked, rendering high-fidelity reports from static markdown content files.
  • Dynamic Sitemap Automation: Wrote a build-time sitemap compiler that maps blog content into discoverable routes for search engines.
// sitemap.ts maps all markdown blog posts dynamically const blogRoutes = posts.map((post) => ({
  url: `https://www.agentstackcalc.com/blog/${post.slug}`,
  lastModified: new Date(post.date),
  changeFrequency: "monthly",
  priority: 0.7
}));
Product Scope
Speech AI Provider cost comparison
LLM Tokens Usage configuration
Automation n8n vs Zapier and Make
Visit Live Platform
Case Study 02

StackVoice TTS Engine

Role: Creator & Applied AI Engineer
Kokoro-82M ONNX FastAPI & Uvicorn Dockerized Container

The Challenge

Managed text-to-speech APIs are convenient, but persistent workloads can make cost, latency, and deployment control important engineering constraints.

The target was a lightweight, CPU-oriented speech API using the open-source Kokoro-82M model, with predictable hosting and direct audio streaming.

Engineering Solution

  • ONNX Runtime Optimization: Used an ONNX representation of Kokoro-82M for efficient CPU inference and a smaller operational footprint.
  • Memory-Cached Stream: Replaced standard file-writing with a non-blocking io.BytesIO structure using Python's soundfile library, sending raw audio bytes directly to the response stream with finalized WAV headers.
  • Docker & Nginx Reverse Proxy: Configured lightweight Debian containers and Nginx reverse proxies with Certbot SSL and custom CORS rules, enabling cross-origin stream queries.
// Non-blocking WAV stream writing using SoundFile context with sf.SoundFile(buffer, mode='w', format='WAV', samplerate=sample_rate, channels=1, subtype='PCM_16') as f:
  f.write(samples)
audio_data = buffer.getvalue()
Key Highlights
<220ms Measured synthesis latency
CPU Hosted Containerized inference
Direct Stream In-memory WAV response
Try Audio Playground
Case Study 03

BidIQ Chrome Extension (AI Assistant)

Role: Creator & Lead Engineer · Jul 2026
Manifest V3 Shadow DOM CSS Secure API Proxy

The Challenge

Freelancers need a faster way to review client signals and prepare a relevant proposal without switching between multiple tools.

BidIQ adds payment-verification checks, average pay-rate context, and AI-assisted proposal drafting directly inside the workflow while keeping the interface isolated from the host page.

Engineering Solution

  • Shadow DOM Encapsulation: Isolated injected interface elements inside a Shadow Root container to prevent most host-page styles from interfering with the companion UI.
  • Manifest V3 background worker: Leveraged asynchronous MV3 service workers to scan page DOM nodes dynamically without slowing down page load performance.
  • Secure Proxy Architecture: Decoupled LLM generation from client-side execution. The extension sends parsed page context to an Express.js backend, which proxies model requests without exposing provider credentials in the browser.
// Injected script creating an isolated Shadow Root container const host = document.createElement('div');
host.id = 'bidiq-root';
document.body.appendChild(host);
const shadowRoot = host.attachShadow({ mode: 'open' });
Product Design
Shadow DOM Interface isolation
API Proxy Server-side model access
MV3 Manifest compliant
View on Chrome Web Store