01 · fresh
Speculative Decoding & DeepSeek DSpark
How modern LLMs generate text up to 3x faster without dropping a single percentage of quality.
12 slices · ~8m
00 / all
The Whole Menu.
Every tile we've ever rolled out: across every counter, on one wall.
counter:
sort:
size:
03 · homemade
Mixture of experts: How AI models got smarter without getting slower
The secret architecture behind trillion-parameter models that stay fast and affordable.
7 slices · ~5m
04 · sauce
The Art of Loop Engineering
How stacking loops turns unpredictable LLMs into production-grade systems.
12 slices · ~9m
01 · fresh
DwarfStar: run frontier AI on your own machine
A hyper-focused, native inference engine built for one job — running DeepSeek V4 locally, even on hardware that technically can't fit it.
12 slices · ~7m
01 · fresh
Figma best practices in the age of AI
AI agents can now read your Figma files directly via MCP. The canvas that worked for human designers breaks for machine consumption. Here is what to change.
13 slices · ~10m
01 · fresh
Open knowledge format: Google's AI memory standard
A minimal YAML spec that turns scattered wikis and PDFs into a structured, Git-trackable knowledge base that LLMs can query without hallucinating the details.
12 slices · ~7m
01 · fresh
Claude Fable 5: The hidden truth
Anthropic's most capable model ever has a secret playbook. The official system card reveals silent classifiers, activation steering, and a 30-day data retention loop you never see.
10 slices · ~7m
01 · fresh
iii: The end of backend fragmentation
A WebSocket-based engine that replaces API gateways, message queues, cron daemons and AI agent scaffolding with one unified runtime.
10 slices · ~6m
01 · fresh
Book-to-Skill: Compile any PDF into native Claude knowledge
An open-source compiler that turns any technical book into a structured Claude Code skill. Pay 4,000 tokens per session instead of 200,000.
10 slices · ~5m
01 · fresh
Agent Memory That Actually Sticks
A fully local engine that cuts token usage by 61% and boosts agent task pass rates by 51%, without calling a single external API.
9 slices · ~5m
01 · fresh
Markdown With Superpowers
One text file. Four output formats. No export pipelines, no tool switching, no compromises. Quarkdown turns plain Markdown into a typesetting language capable of math and logic.
8 slices · ~5m
01 · fresh
Mirage: The Universal File System for AI Agents
How one open-source project gives AI agents a single root directory across every platform, database, and cloud service.
9 slices · ~5m
01 · fresh
When AI Builds Itself: The Dawn of Recursive Self-Improvement
What happens when AI no longer relies on humans to get smarter? The answer is already happening in production labs.
11 slices · ~8m
02 · paper
SkillOpt: Self-Evolving Agent Skills
Microsoft's new system lets AI agents rewrite their own tools when the world changes.
6 slices · ~5m
02 · paper
Attention Is All You Need
The 2017 paper that killed sequential AI and sparked the LLM era.
7 slices · ~6m
02 · paper
BERT & GPT-1: The Fork
The 2018 split that divided AI into readers and writers.
7 slices · ~6m
02 · paper
Language Models are Few-Shot Learners
The 2020 paper that made prompting the new programming.
6 slices · ~5m
02 · paper
InstructGPT & Chinchilla: The Optimization
The two 2022 papers that proved raw scale was only half the answer.
6 slices · ~5m
02 · paper
The Sensory Shift: Native Multimodality
How Gemini 1.5 and GPT-4o taught AI to see, hear, and speak from a single brain.
6 slices · ~5m
02 · paper
The Reasoning Era
How DeepSeek-R1 and s1 moved the AI frontier from training compute to thinking time.
6 slices · ~5m
03 · homemade
Steering vectors: the DJ mixer inside every LLM
Prompting talks to the model. Steering vectors operate inside it.
10 slices · ~7m
03 · homemade
Tokenization: How AI Reads Text
AI doesn't read words. It reads bricks of text called tokens.
8 slices · ~5m
03 · homemade
Embeddings: How AI Understands Meaning
Words as coordinates. The trick behind every AI that feels smart.
8 slices · ~7m
03 · homemade
Vector Databases: Semantic Search at Scale
From keyword matching to finding meaning. The database that understands context.
4 slices · ~5m
03 · homemade
Neural Networks: The Engine of AI
Digital neurons, layered deep. The engine behind every AI breakthrough.
7 slices · ~6m
03 · homemade
Attention Mechanism & Transformers
The architectural secret behind GPT, Claude, and Gemini.
5 slices · ~4m
03 · homemade
Weights & Biases: AI's Control Panel
70 billion dials. All tuned to perfection.
6 slices · ~5m
03 · homemade
Pre-Training vs. Fine-Tuning
Two phases. One brain. Here's how a model goes from clueless to expert.
7 slices · ~5m
03 · homemade
AI Alignment & RLHF
Why brilliant AI can still go terribly wrong. And how we fix it.
6 slices · ~5m
03 · homemade
MCP: The Universal Adapter for AI
One open standard. Every data source. No more custom integration code.
10 slices · ~8m
04 · sauce
The LangChain stack
From raw model to production AI. Here's the full toolkit.
15 slices · ~9m
04 · sauce
Heretic: the abliterated AI models
What happens when AI loses its corporate moral compass? Inside the uncensored mechanics of abliterated neural networks.
11 slices · ~8m
04 · sauce
Caveman
Fewer words. Fewer tokens. Bigger savings.
7 slices · ~5m
04 · sauce
Spec-Driven Context Engineering
Stop context rot. Ship with the GSD Method.
7 slices · ~7m
04 · sauce
The Autonomous Hill-Climber
What if the AI stubbornly fixed its own mistakes until it worked?
8 slices · ~7m
04 · sauce
Prompt Caching
Ask a million times. Pay for one.
7 slices · ~6m
04 · sauce
Progressive Context Disclosure
Give the AI the rules it needs. Nothing more.
8 slices · ~5m