Weekly Digest

The digest that goes out every Monday

Weekly Digest

🍔🧠 How Lovable Cut TTFB 49% After Leaving Next.js

Aug 24, 2026

•

4 min read

🍔🧠 How Lovable Cut TTFB 49% After Leaving Next.js

PLUS: Semantic layer at PostHog 🏗️, AliExpress fingerprinting 🔎 , Git serving rebuilt at Datadog 📈

Alex Zajac

Weekly Digest

🍔🧠 The Weirdest Part of PostHog’s AI Stack Is Just a Markdown File

Aug 17, 2026

•

5 min read

🍔🧠 The Weirdest Part of PostHog’s AI Stack Is Just a Markdown File

PLUS: X ranking algorithm public 👁️, Curl performance tuning ⚡, RAG dispatcher loops 🔄

Alex Zajac

Weekly Digest

Aug 10, 2026

•

4 min read

🍔🧠 What's inside an LLM's KV cache

PLUS: Google's AI depth decline 🧠, Real-time agent load balancing ⚡, Semantic search ranking at scale 👨‍💻

Alex Zajac

Weekly Digest

🍔🧠 Kubernetes Internals: Control Plane Explained

Aug 3, 2026

•

4 min read

🍔🧠 Kubernetes Internals: Control Plane Explained

PLUS: Ray 2x throughput boost ⚡, take-home interview breakdown 🔬, great work principles 📚

Alex Zajac

Weekly Digest

🍔🧠 Software Factories: Harnessing Loops At Scale

Jul 27, 2026

•

5 min read

🍔🧠 Software Factories: Harnessing Loops At Scale

PLUS: In-process load balancing 💨, On-disk ANN indexes 💾, Rust integ tests 🧪

Alex Zajac

Weekly Digest

🍔🧠 Why Linear Feels So Fast (Technical Breakdown)

Jul 20, 2026

•

4 min read

🍔🧠 Why Linear Feels So Fast (Technical Breakdown)

PLUS: Virtual memory explained 👨‍💻, Spark retires vector DBs 💾, Backprop explained simply 🧮

Alex Zajac

Weekly Digest

🍔🧠 The Personalization Stack DoorDash Built (Serves 100M+ Users)

Jul 13, 2026

•

4 min read

🍔🧠 The Personalization Stack DoorDash Built (Serves 100M+ Users)

PLUS: Uniqlo Decoded 🚨, Agentic patterns⚡, Be the idiot mindset 👨‍💻

Alex Zajac

Weekly Digest

🍔🧠 Why Interaction Models Are The Next AI Frontier

Jul 6, 2026

•

4 min read

🍔🧠 Why Interaction Models Are The Next AI Frontier

PLUS: Reddit's anti-spam internals 👨‍💻, p99 0ms autocomplete ⚡, YouTube leaks creators' videos 🚨

Alex Zajac

Weekly Digest

🍔🧠 This DoorDash Engineer Built AI Agents That Run for Days

Jun 29, 2026

•

5 min read

🍔🧠 This DoorDash Engineer Built AI Agents That Run for Days

PLUS: OpenAI's custom chip 🤖, Netflix Cassandra 📊, CQRS deep dive ⚡

Alex Zajac

Weekly Digest

Jun 22, 2026

•

5 min read

🍔🧠 Shopify Said: MySQL > Redis

PLUS: SpaceX buys Cursor for $60B 💰, AWS Blocks framework launch ⚡, Terminal optimization tips 👨‍💻

Alex Zajac

Weekly Digest

🍔🧠 How Stripe Runs a 50M lines Monorepo With 0 Pain

Jun 8, 2026

•

4 min read

🍔🧠 How Stripe Runs a 50M lines Monorepo With 0 Pain

PLUS: Index design fundamentals 👨‍💻, KV performance tool ⚡, Null bugs cost billions 💰

Alex Zajac

Weekly Digest

🍔🧠 All the Distributed Systems Failures in 1 Email

Jun 1, 2026

•

5 min read

🍔🧠 All the Distributed Systems Failures in 1 Email

PLUS: SWE job market 2026 👨‍💻, Visual debugging for ML ⚡, S-tier demo framework 👨‍💻

Alex Zajac

Weekly Digest

🍔🧠 Airbnb's Graph Was So Slow They Rewrote the Engine

May 25, 2026

•

5 min read

🍔🧠 Airbnb's Graph Was So Slow They Rewrote the Engine

PLUS: Vector database deep dive 👨‍💻, Avoiding AI code slop 🤖, DoorDash clusterless ML feature store ⚡

Alex Zajac

Weekly Digest

🍔🧠 Why the F*ck is Discord Still Using Elixir

May 18, 2026

•

5 min read

🍔🧠 Why the F*ck is Discord Still Using Elixir

PLUS: AI as Netflix model 📊, Spotify NLI via Claude 🎵, Subagent patterns 2026 🤖

Alex Zajac

Weekly Digest

🍔🧠 Zero Downtime: How Yelp Upgraded 1000+ Cassandra Nodes (No Rollback)

May 11, 2026

•

5 min read

🍔🧠 Zero Downtime: How Yelp Upgraded 1000+ Cassandra Nodes (No Rollback)

PLUS: OpenAI phone by 2027 📱, Shadow testing Flink 👨‍💻, Computer use 45x expensive 💰

Alex Zajac

Weekly Digest

🍔🧠 How Amazon Uses LLM for Product Recommendation

May 4, 2026

•

6 min read

🍔🧠 How Amazon Uses LLM for Product Recommendation

PLUS: How Shazam Works 🎼 , Agent Harness 🧇, Vercel Breach 🔥

Alex Zajac

Weekly Digest

🍔🧠 How OpenAI Agents Got 40% Faster (System Design)

Apr 27, 2026

•

5 min read

🍔🧠 How OpenAI Agents Got 40% Faster (System Design)

PLUS: DeepSeek's new model 🤖, LLM inference caching 📚, Remote dev setup 👨‍💻

Alex Zajac

Weekly Digest

🍔🧠 NVIDIA Built a Model That Reads 6 Languages at 35 Pages/Second

Apr 20, 2026

•

5 min read

🍔🧠 NVIDIA Built a Model That Reads 6 Languages at 35 Pages/Second

PLUS: Claude Design system 🤖, Netflix ops at live scale ⚡, Data pipeline essentials 👨‍💻

Alex Zajac

Weekly Digest

🍔🧠 How Meta Used AI to Map Tribal Knowledge in Large-Scale Data Pipelines

Apr 13, 2026

•

11 min read

🍔🧠 How Meta Used AI to Map Tribal Knowledge in Large-Scale Data Pipelines

PLUS: Top AI agent benchmarks beaten 🤖, Frontend engineering at Palantir 👨‍💻, Shell tricks productivity ⚡

Alex Zajac

Weekly Digest

🍔🧠 Why Etsy Ditched Keyword Matching for LLMs (40M Queries Later)

Apr 6, 2026

•

11 min read

🍔🧠 Why Etsy Ditched Keyword Matching for LLMs (40M Queries Later)

PLUS: Cursor's new version 📱, Self-healing agents in prod ⚡, JS 2026 guide 👨‍💻

Alex Zajac

Weekly Digest

🍔🧠 How Netflix Reaches 100M Devices in 60 Seconds (Here's the Architecture)

Mar 30, 2026

•

11 min read

🍔🧠 How Netflix Reaches 100M Devices in 60 Seconds (Here's the Architecture)

PLUS: Jensen's future manifesto 🎯, Discord distributed tracing⚡, Async systems + problems 👨‍💻

Alex Zajac

Weekly Digest

🍔🧠 How Facebook Ranks Reels at Scale (Deep Dive)

Mar 23, 2026

•

10 min read

🍔🧠 How Facebook Ranks Reels at Scale (Deep Dive)

PLUS: Astral joins OpenAI 🤖, DeepAgents framework 👨‍💻, WhatsApp architecture deep dive 📱

Alex Zajac

Weekly Digest

🍔🧠 How Spotify Wrapped Works (Real-Time Pattern Matching)

Mar 16, 2026

•

11 min read

🍔🧠 How Spotify Wrapped Works (Real-Time Pattern Matching)

PLUS: System design 114 concepts 👨‍💻, Browser rendering API 🌐, Amazon outage deep dive 🚨

Alex Zajac

Weekly Digest

🍔🧠 Why Discord Didn't Rewrite Elixir (They Built This Instead)

Mar 9, 2026

•

11 min read

🍔🧠 Why Discord Didn't Rewrite Elixir (They Built This Instead)

PLUS: GPT-5.4 arrives 🤖, 60 year old tries Claude Code ⚡, AI investing stack 💰

Alex Zajac

Weekly Digest

🍔🧠 The LLM Architecture That Scales (Mixture of Experts)

Mar 2, 2026

•

11 min read

🍔🧠 The LLM Architecture That Scales (Mixture of Experts)

PLUS: High availability crash course👨‍💻, Next.js rebuilt with AI ⚡, Google's SWE lessons learned 📚

Alex Zajac

Weekly Digest

🍔🧠 How Lovable Cut TTFB 49% After Leaving Next.js

🍔🧠 The Weirdest Part of PostHog’s AI Stack Is Just a Markdown File

🍔🧠 What's inside an LLM's KV cache

🍔🧠 Kubernetes Internals: Control Plane Explained

🍔🧠 Software Factories: Harnessing Loops At Scale

🍔🧠 Why Linear Feels So Fast (Technical Breakdown)

🍔🧠 The Personalization Stack DoorDash Built (Serves 100M+ Users)

🍔🧠 Why Interaction Models Are The Next AI Frontier

🍔🧠 This DoorDash Engineer Built AI Agents That Run for Days

🍔🧠 Shopify Said: MySQL > Redis

🍔🧠 How Stripe Runs a 50M lines Monorepo With 0 Pain

🍔🧠 All the Distributed Systems Failures in 1 Email

🍔🧠 Airbnb's Graph Was So Slow They Rewrote the Engine

🍔🧠 Why the F*ck is Discord Still Using Elixir

🍔🧠 Zero Downtime: How Yelp Upgraded 1000+ Cassandra Nodes (No Rollback)

🍔🧠 How Amazon Uses LLM for Product Recommendation

🍔🧠 How OpenAI Agents Got 40% Faster (System Design)

🍔🧠 NVIDIA Built a Model That Reads 6 Languages at 35 Pages/Second

🍔🧠 How Meta Used AI to Map Tribal Knowledge in Large-Scale Data Pipelines

🍔🧠 Why Etsy Ditched Keyword Matching for LLMs (40M Queries Later)

🍔🧠 How Netflix Reaches 100M Devices in 60 Seconds (Here's the Architecture)

🍔🧠 How Facebook Ranks Reels at Scale (Deep Dive)

🍔🧠 How Spotify Wrapped Works (Real-Time Pattern Matching)

🍔🧠 Why Discord Didn't Rewrite Elixir (They Built This Instead)

🍔🧠 The LLM Architecture That Scales (Mixture of Experts)

Hungry Minds 🍔🧠