ALL Blogs

SnackOnAI Blog

Handy: The Most Forkable Speech-to-Text App Is a Better Design Goal Than the Most Accurate One

Jul 6, 2026

•

12 min read

Handy: The Most Forkable Speech-to-Text App Is a Better Design Goal Than the Most Accurate One

The author of Handy (cjpais/Handy, MIT, 21k stars, v0.8.3) wrote this explicitly: "Handy isn't trying to be the best speech-to-text app, it's trying to be the most forkable one."

Mohinish S

SnackOnAI Blog

Meetily: The AI Meeting Assistant That Proves Privacy and Cloud Are Not a Tradeoff. They Are an Architecture Decision.

Jul 5, 2026

•

12 min read

Meetily: The AI Meeting Assistant That Proves Privacy and Cloud Are Not a Tradeoff. They Are an Architecture Decision.

Every AI meeting tool you have seen evaluated sends your recordings to someone else's server. Meetily (Zackriya-Solutions/meetily, MIT, 15.4k stars) does not. It is a self-contained Tauri desktop application that captures audio, transcribes with Whisper or Parakeet, and summarizes via Ollama or any OpenAI-compatible endpoint, entirely on your local machine.

Mohinish S

SnackOnAI Blog

Loop Engineering: The Claude Code Paper Proves Your Agent Loop Is the Smallest Part. The Systems Around It Are Where the Real Engineering Lives.

Jul 4, 2026

•

13 min read

Loop Engineering: The Claude Code Paper Proves Your Agent Loop Is the Smallest Part. The Systems Around It Are Where the Real Engineering Lives.

Nine out of ten developers still prompt their AI coding agents by hand. For those who have moved to automated loops, the mistake is treating the loop itself as the engineering problem.

Mohinish S

SnackOnAI Blog

Hindsight: Agent Memory That Lifts a 20B Open-Source Model From 39% to 83.6% on Long-Horizon Benchmarks, Past Full-Context GPT-4o

Jul 2, 2026

•

14 min read

Hindsight: Agent Memory That Lifts a 20B Open-Source Model From 39% to 83.6% on Long-Horizon Benchmarks, Past Full-Context GPT-4o

Every agent memory system built in 2025 made the same mistake: treating memory as a retrieval problem. Dump facts into a vector store, fetch top-k, inject into context.

Mohinish S

SnackOnAI Blog

LuxTTS: The 48kHz Voice Cloning Model That Fits in 1GB of VRAM and Runs 150x Faster Than Real Time Is Built on an ASR Architecture Nobody Expected to Work for TTS

Jul 1, 2026

•

12 min read

LuxTTS: The 48kHz Voice Cloning Model That Fits in 1GB of VRAM and Runs 150x Faster Than Real Time Is Built on an ASR Architecture Nobody Expected to Work for TTS

Every production TTS system in 2025 makes the same engineering tradeoff: quality requires a large diffusion model with many inference steps, and speed requires sacrificing either.

Mohinish S

SnackOnAI Blog

PixelRAG: Your RAG Pipeline Is Losing 40% of the Evidence Before the LLM Ever Sees It. The Fix Is Screenshots.

Jun 30, 2026

•

13 min read

PixelRAG: Your RAG Pipeline Is Losing 40% of the Evidence Before the LLM Ever Sees It. The Fix Is Screenshots.

HTML parsers are the silent quality tax on every RAG pipeline. They discard 40%+ of recoverable text from a web page, and choosing a different parser changes your final accuracy by up to 10 percentage points on SimpleQA.

Mohinish S

ALL Blogs

Handy: The Most Forkable Speech-to-Text App Is a Better Design Goal Than the Most Accurate One

Meetily: The AI Meeting Assistant That Proves Privacy and Cloud Are Not a Tradeoff. They Are an Architecture Decision.

Loop Engineering: The Claude Code Paper Proves Your Agent Loop Is the Smallest Part. The Systems Around It Are Where the Real Engineering Lives.

Hindsight: Agent Memory That Lifts a 20B Open-Source Model From 39% to 83.6% on Long-Horizon Benchmarks, Past Full-Context GPT-4o

LuxTTS: The 48kHz Voice Cloning Model That Fits in 1GB of VRAM and Runs 150x Faster Than Real Time Is Built on an ASR Architecture Nobody Expected to Work for TTS

PixelRAG: Your RAG Pipeline Is Losing 40% of the Evidence Before the LLM Ever Sees It. The Fix Is Screenshots.

Quick Links

Subscription

Socials