Forem: H33.ai

Building a Sub-Microsecond Cache for a Billion-User Mining Platform

H33.ai — Thu, 16 Apr 2026 01:57:22 +0000

The Problem: 100-250ms Middleware Tax

Our Node.js Express backend had a dirty secret: before any route handler ran, the middleware stack consumed 100-250ms. Four separate Redis round-trips for rate limiting, session load, XSS sanitization — all serial.

For a platform targeting billions of users, this was unacceptable.

The Solution: CacheeEngine in Rust

We replaced the entire JS middleware stack with an in-process Rust cache engine called CacheeEngine:

Component	What It Does	Memory
CacheeLFU eviction	Frequency counters with periodic halving decay	Inline
Count-Min Sketch	Admission doorkeeper — rejects one-hit-wonders	512 KiB constant
DashMap storage	Lock-free concurrent reads	O(entries)
SWR support	Stale-while-revalidate built in	Built-in

CacheeLFU vs W-TinyLFU

We chose CacheeLFU over W-TinyLFU (Caffeine/moka):

No window cache
Direct frequency comparison with CMS admission
Periodic halving decay instead of reset-on-epoch

Count-Min Sketch: 512 KiB That Saves Everything

The CMS is the doorkeeper. Layout: 4 rows x 131,072 counters x 1 byte = 512 KiB. Constant regardless of entry count.

Results

Path	JS (Express)	Rust (Axum + Cachee)
Middleware	100-250ms	<5ms
Trust score hit	5-10ms	<1us
Mining sync	50ms-2s	<10ms
Swap quote stale	40-150ms	<1us
Rate limit	30-50ms	<100ns

The binary is 5.2MB stripped. 15 tests pass.

RevMine is live. Check GitHub for open-source components.

Built on Cachee — H33 post-quantum cache engine with CacheeLFU eviction.

Why Loyalty Points Are Broken (And What Revenue-Backed Tokens Fix)

H33.ai — Thu, 16 Apr 2026 01:57:16 +0000

The $300B Loyalty Problem

Traditional loyalty points have a 54% abandonment rate. Customers earn points, forget about them, and churn.

What if loyalty points were actually worth something?

Revenue-Backed Tokens

RevMine replaces static points with revenue-backed Solana tokens. When a business grows, the tokens grow. This creates real switching costs — 25-40% churn reduction measured across early adopters.

How It Works

Business embeds a widget (2 lines of code)
Users mine tokens by engaging
Mining rates are AI-optimized
Tokens backed by revenue via Stripe verification
On-chain settlement on Solana

The Tech

The backend is 100% Rust:

Sub-15ms API latency (Axum + tokio)
CacheeLFU in-process cache with Count-Min Sketch (512 KiB constant)
Post-quantum encryption via H33/NIST FIPS 204
Offline-first WASM mining engine
Single SQL transaction for mining validation

For SaaS Founders

If you are fighting churn:

Free tier, no risk
5-minute setup, widget embed
No crypto knowledge required for users
White-label, your brand

Try the token wizard to see token creation.

Built with Rust, Solana, and Cachee post-quantum caching.

20 Rust Microservices, Zero Node.js: How We Rebuilt Our Entire Video Platform

H33.ai — Thu, 16 Apr 2026 01:31:22 +0000

At V100, we build AI video infrastructure entirely in Rust. 20 microservices. 0.01ms server processing. 220,000+ requests per second. Post-quantum encryption on every call.

This post is a deep dive into our architecture and the engineering decisions behind it.

Why Rust for Video Infrastructure?

Video infrastructure has unique constraints that make Rust the ideal choice:

Zero GC pauses — Real-time video processing can't tolerate garbage collection stops
Memory safety — Buffer overflows in media pipelines are a security nightmare
Performance — Our gateway processes requests in 10 microseconds, not 10 milliseconds
Small binaries — Our meeting signaling server is a 2MB binary serving WebRTC at scale

Our Stack

Component	Technology
Web Framework	Axum + Tokio
Database	PostgreSQL + TimescaleDB
Cache	Redis + Cachee (in-process)
Media	FFmpeg (sidecar)
Crypto	ML-KEM-768 + ML-DSA-65 (post-quantum)
Infra	Docker + AWS ECS Fargate

The Numbers

20 Rust microservices — gateway, AI, transcription, video, conferencing, billing, broadcasting
0.01ms server processing latency
220K+ RPS sustained throughput
263ns pipeline latency
938 tests with zero failures
40+ languages for real-time transcription
7 platforms for social publishing (YouTube, TikTok, Instagram, LinkedIn, X, Facebook, Vimeo)

Key Services

Gateway (Axum) — JWT auth, rate limiting, CSRF/SSRF protection
AI Orchestration — Claude/Gemini proxy, streaming, compliance
Transcription — Deepgram + Whisper, word-level timestamps, 40+ languages
Meeting Signaling — WebRTC SDP/ICE, DashMap concurrency, 2MB binary
v100-turn — Full broadcast platform: ABR, DRM, DVR, spatial audio, deepfake detection

Try It

V100 has a free tier — v100.ai/pricing

API docs at docs.v100.ai

Read the full technical deep-dive at v100.ai/blog/20-rust-microservices-zero-nodejs

V100 is built by H33.ai — post-quantum security for production systems.

10 Microseconds: How We Built the Fastest API Gateway in Rust

H33.ai — Thu, 16 Apr 2026 01:31:20 +0000

At V100, we build AI video infrastructure entirely in Rust. 20 microservices. 0.01ms server processing. 220,000+ requests per second. Post-quantum encryption on every call.

This post is a deep dive into our architecture and the engineering decisions behind it.

Why Rust for Video Infrastructure?

Video infrastructure has unique constraints that make Rust the ideal choice:

Zero GC pauses — Real-time video processing can't tolerate garbage collection stops
Memory safety — Buffer overflows in media pipelines are a security nightmare
Performance — Our gateway processes requests in 10 microseconds, not 10 milliseconds
Small binaries — Our meeting signaling server is a 2MB binary serving WebRTC at scale

Our Stack

Component	Technology
Web Framework	Axum + Tokio
Database	PostgreSQL + TimescaleDB
Cache	Redis + Cachee (in-process)
Media	FFmpeg (sidecar)
Crypto	ML-KEM-768 + ML-DSA-65 (post-quantum)
Infra	Docker + AWS ECS Fargate

The Numbers

20 Rust microservices — gateway, AI, transcription, video, conferencing, billing, broadcasting
0.01ms server processing latency
220K+ RPS sustained throughput
263ns pipeline latency
938 tests with zero failures
40+ languages for real-time transcription
7 platforms for social publishing (YouTube, TikTok, Instagram, LinkedIn, X, Facebook, Vimeo)

Key Services

Gateway (Axum) — JWT auth, rate limiting, CSRF/SSRF protection
AI Orchestration — Claude/Gemini proxy, streaming, compliance
Transcription — Deepgram + Whisper, word-level timestamps, 40+ languages
Meeting Signaling — WebRTC SDP/ICE, DashMap concurrency, 2MB binary
v100-turn — Full broadcast platform: ABR, DRM, DVR, spatial audio, deepfake detection

Try It

V100 has a free tier — v100.ai/pricing

API docs at docs.v100.ai

Read the full technical deep-dive at v100.ai/blog/10-microsecond-api-gateway

V100 is built by H33.ai — post-quantum security for production systems.

Rust vs C++ for Video Server Performance: Why We Chose Rust

H33.ai — Thu, 16 Apr 2026 01:31:19 +0000

At V100, we build AI video infrastructure entirely in Rust. 20 microservices. 0.01ms server processing. 220,000+ requests per second. Post-quantum encryption on every call.

This post is a deep dive into our architecture and the engineering decisions behind it.

Why Rust for Video Infrastructure?

Video infrastructure has unique constraints that make Rust the ideal choice:

Zero GC pauses — Real-time video processing can't tolerate garbage collection stops
Memory safety — Buffer overflows in media pipelines are a security nightmare
Performance — Our gateway processes requests in 10 microseconds, not 10 milliseconds
Small binaries — Our meeting signaling server is a 2MB binary serving WebRTC at scale

Our Stack

Component	Technology
Web Framework	Axum + Tokio
Database	PostgreSQL + TimescaleDB
Cache	Redis + Cachee (in-process)
Media	FFmpeg (sidecar)
Crypto	ML-KEM-768 + ML-DSA-65 (post-quantum)
Infra	Docker + AWS ECS Fargate

The Numbers

20 Rust microservices — gateway, AI, transcription, video, conferencing, billing, broadcasting
0.01ms server processing latency
220K+ RPS sustained throughput
263ns pipeline latency
938 tests with zero failures
40+ languages for real-time transcription
7 platforms for social publishing (YouTube, TikTok, Instagram, LinkedIn, X, Facebook, Vimeo)

Key Services

Gateway (Axum) — JWT auth, rate limiting, CSRF/SSRF protection
AI Orchestration — Claude/Gemini proxy, streaming, compliance
Transcription — Deepgram + Whisper, word-level timestamps, 40+ languages
Meeting Signaling — WebRTC SDP/ICE, DashMap concurrency, 2MB binary
v100-turn — Full broadcast platform: ABR, DRM, DVR, spatial audio, deepfake detection

Try It

V100 has a free tier — v100.ai/pricing

API docs at docs.v100.ai

Read the full technical deep-dive at v100.ai/blog/rust-vs-cpp-video-server-performance

V100 is built by H33.ai — post-quantum security for production systems.

Fastest WebRTC Server in 2026: Benchmarks Against coturn and LiveKit

H33.ai — Thu, 16 Apr 2026 01:31:17 +0000

At V100, we build AI video infrastructure entirely in Rust. 20 microservices. 0.01ms server processing. 220,000+ requests per second. Post-quantum encryption on every call.

This post is a deep dive into our architecture and the engineering decisions behind it.

Why Rust for Video Infrastructure?

Video infrastructure has unique constraints that make Rust the ideal choice:

Zero GC pauses — Real-time video processing can't tolerate garbage collection stops
Memory safety — Buffer overflows in media pipelines are a security nightmare
Performance — Our gateway processes requests in 10 microseconds, not 10 milliseconds
Small binaries — Our meeting signaling server is a 2MB binary serving WebRTC at scale

Our Stack

Component	Technology
Web Framework	Axum + Tokio
Database	PostgreSQL + TimescaleDB
Cache	Redis + Cachee (in-process)
Media	FFmpeg (sidecar)
Crypto	ML-KEM-768 + ML-DSA-65 (post-quantum)
Infra	Docker + AWS ECS Fargate

The Numbers

20 Rust microservices — gateway, AI, transcription, video, conferencing, billing, broadcasting
0.01ms server processing latency
220K+ RPS sustained throughput
263ns pipeline latency
938 tests with zero failures
40+ languages for real-time transcription
7 platforms for social publishing (YouTube, TikTok, Instagram, LinkedIn, X, Facebook, Vimeo)

Key Services

Gateway (Axum) — JWT auth, rate limiting, CSRF/SSRF protection
AI Orchestration — Claude/Gemini proxy, streaming, compliance
Transcription — Deepgram + Whisper, word-level timestamps, 40+ languages
Meeting Signaling — WebRTC SDP/ICE, DashMap concurrency, 2MB binary
v100-turn — Full broadcast platform: ABR, DRM, DVR, spatial audio, deepfake detection

Try It

V100 has a free tier — v100.ai/pricing

API docs at docs.v100.ai

Read the full technical deep-dive at v100.ai/blog/fastest-webrtc-server-2026

V100 is built by H33.ai — post-quantum security for production systems.

Sub-Microsecond Video Processing: Inside Our 263ns Pipeline

H33.ai — Thu, 16 Apr 2026 01:31:16 +0000

At V100, we build AI video infrastructure entirely in Rust. 20 microservices. 0.01ms server processing. 220,000+ requests per second. Post-quantum encryption on every call.

This post is a deep dive into our architecture and the engineering decisions behind it.

Why Rust for Video Infrastructure?

Video infrastructure has unique constraints that make Rust the ideal choice:

Zero GC pauses — Real-time video processing can't tolerate garbage collection stops
Memory safety — Buffer overflows in media pipelines are a security nightmare
Performance — Our gateway processes requests in 10 microseconds, not 10 milliseconds
Small binaries — Our meeting signaling server is a 2MB binary serving WebRTC at scale

Our Stack

Component	Technology
Web Framework	Axum + Tokio
Database	PostgreSQL + TimescaleDB
Cache	Redis + Cachee (in-process)
Media	FFmpeg (sidecar)
Crypto	ML-KEM-768 + ML-DSA-65 (post-quantum)
Infra	Docker + AWS ECS Fargate

The Numbers

20 Rust microservices — gateway, AI, transcription, video, conferencing, billing, broadcasting
0.01ms server processing latency
220K+ RPS sustained throughput
263ns pipeline latency
938 tests with zero failures
40+ languages for real-time transcription
7 platforms for social publishing (YouTube, TikTok, Instagram, LinkedIn, X, Facebook, Vimeo)

Key Services

Gateway (Axum) — JWT auth, rate limiting, CSRF/SSRF protection
AI Orchestration — Claude/Gemini proxy, streaming, compliance
Transcription — Deepgram + Whisper, word-level timestamps, 40+ languages
Meeting Signaling — WebRTC SDP/ICE, DashMap concurrency, 2MB binary
v100-turn — Full broadcast platform: ABR, DRM, DVR, spatial audio, deepfake detection

Try It

V100 has a free tier — v100.ai/pricing

API docs at docs.v100.ai

Read the full technical deep-dive at v100.ai/blog/sub-microsecond-video-processing

V100 is built by H33.ai — post-quantum security for production systems.

31 Nanoseconds: How Cachee Powers the Fastest Video API Cache

H33.ai — Thu, 16 Apr 2026 01:31:14 +0000

At V100, we build AI video infrastructure entirely in Rust. 20 microservices. 0.01ms server processing. 220,000+ requests per second. Post-quantum encryption on every call.

This post is a deep dive into our architecture and the engineering decisions behind it.

Why Rust for Video Infrastructure?

Video infrastructure has unique constraints that make Rust the ideal choice:

Zero GC pauses — Real-time video processing can't tolerate garbage collection stops
Memory safety — Buffer overflows in media pipelines are a security nightmare
Performance — Our gateway processes requests in 10 microseconds, not 10 milliseconds
Small binaries — Our meeting signaling server is a 2MB binary serving WebRTC at scale

Our Stack

Component	Technology
Web Framework	Axum + Tokio
Database	PostgreSQL + TimescaleDB
Cache	Redis + Cachee (in-process)
Media	FFmpeg (sidecar)
Crypto	ML-KEM-768 + ML-DSA-65 (post-quantum)
Infra	Docker + AWS ECS Fargate

The Numbers

20 Rust microservices — gateway, AI, transcription, video, conferencing, billing, broadcasting
0.01ms server processing latency
220K+ RPS sustained throughput
263ns pipeline latency
938 tests with zero failures
40+ languages for real-time transcription
7 platforms for social publishing (YouTube, TikTok, Instagram, LinkedIn, X, Facebook, Vimeo)

Key Services

Gateway (Axum) — JWT auth, rate limiting, CSRF/SSRF protection
AI Orchestration — Claude/Gemini proxy, streaming, compliance
Transcription — Deepgram + Whisper, word-level timestamps, 40+ languages
Meeting Signaling — WebRTC SDP/ICE, DashMap concurrency, 2MB binary
v100-turn — Full broadcast platform: ABR, DRM, DVR, spatial audio, deepfake detection

Try It

V100 has a free tier — v100.ai/pricing

API docs at docs.v100.ai

Read the full technical deep-dive at v100.ai/blog/cachee-31-nanosecond-cache

V100 is built by H33.ai — post-quantum security for production systems.

How We Built Post-Quantum Encrypted Video Conferencing

H33.ai — Thu, 16 Apr 2026 01:31:12 +0000

At V100, we build AI video infrastructure entirely in Rust. 20 microservices. 0.01ms server processing. 220,000+ requests per second. Post-quantum encryption on every call.

This post is a deep dive into our architecture and the engineering decisions behind it.

Why Rust for Video Infrastructure?

Video infrastructure has unique constraints that make Rust the ideal choice:

Zero GC pauses — Real-time video processing can't tolerate garbage collection stops
Memory safety — Buffer overflows in media pipelines are a security nightmare
Performance — Our gateway processes requests in 10 microseconds, not 10 milliseconds
Small binaries — Our meeting signaling server is a 2MB binary serving WebRTC at scale

Our Stack

Component	Technology
Web Framework	Axum + Tokio
Database	PostgreSQL + TimescaleDB
Cache	Redis + Cachee (in-process)
Media	FFmpeg (sidecar)
Crypto	ML-KEM-768 + ML-DSA-65 (post-quantum)
Infra	Docker + AWS ECS Fargate

The Numbers

20 Rust microservices — gateway, AI, transcription, video, conferencing, billing, broadcasting
0.01ms server processing latency
220K+ RPS sustained throughput
263ns pipeline latency
938 tests with zero failures
40+ languages for real-time transcription
7 platforms for social publishing (YouTube, TikTok, Instagram, LinkedIn, X, Facebook, Vimeo)

Key Services

Gateway (Axum) — JWT auth, rate limiting, CSRF/SSRF protection
AI Orchestration — Claude/Gemini proxy, streaming, compliance
Transcription — Deepgram + Whisper, word-level timestamps, 40+ languages
Meeting Signaling — WebRTC SDP/ICE, DashMap concurrency, 2MB binary
v100-turn — Full broadcast platform: ABR, DRM, DVR, spatial audio, deepfake detection

Try It

V100 has a free tier — v100.ai/pricing

API docs at docs.v100.ai

Read the full technical deep-dive at v100.ai/blog/post-quantum-encrypted-meetings

V100 is built by H33.ai — post-quantum security for production systems.

Open-Sourcing RustTURN: Post-Quantum Video Infrastructure

H33.ai — Thu, 16 Apr 2026 01:31:11 +0000

At V100, we build AI video infrastructure entirely in Rust. 20 microservices. 0.01ms server processing. 220,000+ requests per second. Post-quantum encryption on every call.

This post is a deep dive into our architecture and the engineering decisions behind it.

Why Rust for Video Infrastructure?

Video infrastructure has unique constraints that make Rust the ideal choice:

Zero GC pauses — Real-time video processing can't tolerate garbage collection stops
Memory safety — Buffer overflows in media pipelines are a security nightmare
Performance — Our gateway processes requests in 10 microseconds, not 10 milliseconds
Small binaries — Our meeting signaling server is a 2MB binary serving WebRTC at scale

Our Stack

Component	Technology
Web Framework	Axum + Tokio
Database	PostgreSQL + TimescaleDB
Cache	Redis + Cachee (in-process)
Media	FFmpeg (sidecar)
Crypto	ML-KEM-768 + ML-DSA-65 (post-quantum)
Infra	Docker + AWS ECS Fargate

The Numbers

20 Rust microservices — gateway, AI, transcription, video, conferencing, billing, broadcasting
0.01ms server processing latency
220K+ RPS sustained throughput
263ns pipeline latency
938 tests with zero failures
40+ languages for real-time transcription
7 platforms for social publishing (YouTube, TikTok, Instagram, LinkedIn, X, Facebook, Vimeo)

Key Services

Gateway (Axum) — JWT auth, rate limiting, CSRF/SSRF protection
AI Orchestration — Claude/Gemini proxy, streaming, compliance
Transcription — Deepgram + Whisper, word-level timestamps, 40+ languages
Meeting Signaling — WebRTC SDP/ICE, DashMap concurrency, 2MB binary
v100-turn — Full broadcast platform: ABR, DRM, DVR, spatial audio, deepfake detection

Try It

V100 has a free tier — v100.ai/pricing

API docs at docs.v100.ai

Read the full technical deep-dive at v100.ai/blog/open-sourcing-rustturn

V100 is built by H33.ai — post-quantum security for production systems.

938 Tests, Zero Failures: Why We're the Most Tested Video API

H33.ai — Thu, 16 Apr 2026 01:31:09 +0000

At V100, we build AI video infrastructure entirely in Rust. 20 microservices. 0.01ms server processing. 220,000+ requests per second. Post-quantum encryption on every call.

This post is a deep dive into our architecture and the engineering decisions behind it.

Why Rust for Video Infrastructure?

Video infrastructure has unique constraints that make Rust the ideal choice:

Zero GC pauses — Real-time video processing can't tolerate garbage collection stops
Memory safety — Buffer overflows in media pipelines are a security nightmare
Performance — Our gateway processes requests in 10 microseconds, not 10 milliseconds
Small binaries — Our meeting signaling server is a 2MB binary serving WebRTC at scale

Our Stack

Component	Technology
Web Framework	Axum + Tokio
Database	PostgreSQL + TimescaleDB
Cache	Redis + Cachee (in-process)
Media	FFmpeg (sidecar)
Crypto	ML-KEM-768 + ML-DSA-65 (post-quantum)
Infra	Docker + AWS ECS Fargate

The Numbers

20 Rust microservices — gateway, AI, transcription, video, conferencing, billing, broadcasting
0.01ms server processing latency
220K+ RPS sustained throughput
263ns pipeline latency
938 tests with zero failures
40+ languages for real-time transcription
7 platforms for social publishing (YouTube, TikTok, Instagram, LinkedIn, X, Facebook, Vimeo)

Key Services

Gateway (Axum) — JWT auth, rate limiting, CSRF/SSRF protection
AI Orchestration — Claude/Gemini proxy, streaming, compliance
Transcription — Deepgram + Whisper, word-level timestamps, 40+ languages
Meeting Signaling — WebRTC SDP/ICE, DashMap concurrency, 2MB binary
v100-turn — Full broadcast platform: ABR, DRM, DVR, spatial audio, deepfake detection

Try It

V100 has a free tier — v100.ai/pricing

API docs at docs.v100.ai

Read the full technical deep-dive at v100.ai/blog/938-tests-zero-failures

V100 is built by H33.ai — post-quantum security for production systems.

Real-Time Video Intelligence: Our AI Pipeline Running at 220K RPS

H33.ai — Thu, 16 Apr 2026 01:31:07 +0000

At V100, we build AI video infrastructure entirely in Rust. 20 microservices. 0.01ms server processing. 220,000+ requests per second. Post-quantum encryption on every call.

This post is a deep dive into our architecture and the engineering decisions behind it.

Why Rust for Video Infrastructure?

Video infrastructure has unique constraints that make Rust the ideal choice:

Zero GC pauses — Real-time video processing can't tolerate garbage collection stops
Memory safety — Buffer overflows in media pipelines are a security nightmare
Performance — Our gateway processes requests in 10 microseconds, not 10 milliseconds
Small binaries — Our meeting signaling server is a 2MB binary serving WebRTC at scale

Our Stack

Component	Technology
Web Framework	Axum + Tokio
Database	PostgreSQL + TimescaleDB
Cache	Redis + Cachee (in-process)
Media	FFmpeg (sidecar)
Crypto	ML-KEM-768 + ML-DSA-65 (post-quantum)
Infra	Docker + AWS ECS Fargate

The Numbers

20 Rust microservices — gateway, AI, transcription, video, conferencing, billing, broadcasting
0.01ms server processing latency
220K+ RPS sustained throughput
263ns pipeline latency
938 tests with zero failures
40+ languages for real-time transcription
7 platforms for social publishing (YouTube, TikTok, Instagram, LinkedIn, X, Facebook, Vimeo)

Key Services

Gateway (Axum) — JWT auth, rate limiting, CSRF/SSRF protection
AI Orchestration — Claude/Gemini proxy, streaming, compliance
Transcription — Deepgram + Whisper, word-level timestamps, 40+ languages
Meeting Signaling — WebRTC SDP/ICE, DashMap concurrency, 2MB binary
v100-turn — Full broadcast platform: ABR, DRM, DVR, spatial audio, deepfake detection

Try It

V100 has a free tier — v100.ai/pricing

API docs at docs.v100.ai

Read the full technical deep-dive at v100.ai/blog/real-time-video-intelligence-220k-rps

V100 is built by H33.ai — post-quantum security for production systems.