-
Notifications
You must be signed in to change notification settings - Fork 0
Oct
Maciej J edited this page Aug 24, 2026
·
141 revisions
128 links - Work 107 · Music 1 · Fun 8 · Life 12
On this page: Work · Music · Fun · Life
- A ChainForge - Visual Programming for Prompt Experimentation - Open-source visual environment for designing, testing, iterating on prompts across multiple LLMs. · #prompt-testing #visualization #llm-tools
- A OpenPipe - Convert Expensive Prompts to Fine-Tuned Models - Tool for optimizing LLM costs by converting expensive prompts into efficient fine-tuned models. · #llm-optimization #cost-reduction #fine-tuning
- A Playwright Skill - Claude Code Browser Automation - Claude Code skill enabling autonomous browser automation testing and validation via Playwright. · #automation #browser-testing #claude-skills
- A Living dangerously with Claude - Talk on autonomy tradeoffs in coding agents, discussing risks and benefits of agentic control. · #llm-agents #claude-code #ai-safety
- A Introducing Claude Skills - Anthropic's announcement of Skills—new pattern for extending Claude with custom capabilities. · #claude #agent-skills
- A nanochat: ChatGPT Implementation for $100 - Full ChatGPT-style LLM including training, inference, and web UI in minimal code by Karpathy. · #llm-implementation #ai-tools
- A The Missing Semester of Your CS Education - MIT course teaching practical developer tools and productivity techniques beyond algorithms. · #education #developer-tools #productivity · M
- A code2prompt - Codebase to LLM Prompt - CLI tool that packages your entire codebase into optimized LLM prompts with token counting. · #llm-tools #prompt-eng
- A 90%: AI Now Writing Most of My Code - Armin Ronacher's reflections on using AI agents for most practical daily programming work. · #coding-agent #workflow
- A Bringing memory to Claude - Memory built for work - Anthropic announcement of Claude's persistent memory feature for team workflow continuity. · #llm-features #claude-api #agent-memory
- A Anthropic's Interactive Prompt Engineering Tutorial - Official interactive tutorial teaching prompt engineering best practices from Anthropic. · #learning #ai-tools
- A 12-Factor Agents - Production LLM Design - Foundational principles for building robust, production-ready software powered by LLMs. · #llm-agents #best-practices
- A superpowers: Claude Code Agentic Skills Framework - Core methodology and skills library for effective AI-assisted software development. · #coding-agent #workflow
- A Claude Skills Bigger Than MCP - Simon Willison's analysis of why Claude Skills may be more significant than MCP. · #claude #agent-skills
- A Claude Haiku 4.5 Release - Anthropic's announcement of Claude Haiku 4.5—latest small model for all users. · #claude #llm-models
- A Haiku 4.5: Sonnet Performance at 3x Cheaper - Analysis of Claude Haiku 4.5 performance benchmarks and cost advantages. · #claude #llm-models
- B Learn Prompting - Prompt Engineering Guide - Comprehensive GitHub resource with tutorials on prompt engineering, generative AI, and LLM fundamentals for beginners. · #prompt-eng #llm-learning
- B Stable Code 3B – Coding Model for Ollama - Compact 3B coding model matching Code Llama 7B performance; instruct and completion variants for local deployment. · #llm #code-generation #local-ai
- B The 28 AI Tools I Wish Existed - Sharif Shameem identifies gaps in AI tooling ecosystem; product strategy and design thinking for AI companies. · #ai-products #tool-design #product-strategy
- B After Nine Years, Replit Found Its Market - TechCrunch analysis of Replit's $3B path versus AI coding startups; business resilience and market timing. · #startup #ai-coding #business-strategy
- B Awesome Prompt Engineering - Hand-curated collection of prompt engineering resources focused on GPT, ChatGPT, and generative AI models. · #prompt-eng #collections
- B Moondream 2 – Vision Language Model - Small VLM designed for efficient edge device deployment; integrates vision understanding with language. · #llm #vision #edge-ai
- B SmoLLM – Small Language Model Family - 135M–1.7B parameter models trained on high-quality data; practical for resource-constrained inference. · #llm #small-models #training-data
- B StableLM Zephyr – Lightweight Chat Model - Fast chat model optimized for edge devices; accurate responses without high-end hardware requirements. · #llm #chat #edge-ai
- B Extract-0: Specialized LLM for Document Information Extraction - 7B parameter model optimized for document extraction; exceeds much larger models on extraction tasks. · #nlp #document-extraction #llm-research
- B PromptBench - LLM Evaluation Framework - Unified evaluation framework for systematically testing and benchmarking language model performance. · #llm-evaluation #benchmarking #testing
- B Code like a surgeon - Essay exploring surgical precision analogy in programming emphasizing deliberate coding practices. · #programming #code-quality #craftsmanship
- B Sandbox Runtime - Anthropic Process Sandboxing - Lightweight OS-level sandboxing tool restricting filesystem and network access without containers. · #sandbox #security #systems
- B Smallthinker – Small Reasoning Model - 3B reasoning model fine-tuned from Qwen 2.5; adds chain-of-thought capability to edge deployments. · #llm #reasoning #small-models
- B The Mad King's Digital Killswitch - Pluralistic - Cory Doctorow analysis of digital infrastructure control, censorship, and internet fragmentation. · #internet-policy #censorship #tech-ethics
- B Aim - ML Experiment Tracker - Open-source platform for tracking, comparing, and managing machine learning experiment metadata. · #ml-ops #experiment-tracking
-
B Creating a Vector Database for RAG - Tutorial on building and optimizing vector databases for retrieval-augmented generation systems. · #rag #vector-database ·
dead - B ell - Language Model Programming Library - Python library for building structured, composable programs using language models as primitives. · #llm-libraries #python
- B Prompt Engineering - 22 Techniques - Repository of 22 prompt engineering methods with executable Jupyter notebooks from fundamentals. · #prompt-eng #tutorials
- B Swarms AI - Multi-Agent Framework - Enterprise-grade orchestration platform for building, deploying, and scaling multiple AI agents. · #llm-agents #frameworks
- B "Fuck You" Companies: Be Different First, Then Better - Essay on startup differentiation focusing on unique positioning before incremental improvement. · #startup-strategy #business
- B PromptTools - Prompt Testing & Experimentation Framework - Open-source toolkit supporting prompt testing, LLM evaluation, and vector database integration. · #prompt-testing #llm-tools #vector-db
- B Intuitive RL: Intro to Advantage-Actor-Critic (A2C) - Accessible introduction to reinforcement learning using animal learning as conceptual analogy. · #reinforcement-learning #ml-algorithms #learning
- B Learning Prompt - Free Course - Free prompt engineering course with hands-on tutorials for ChatGPT and Midjourney integration. · #prompt-eng #courses
- B Deloitte Refunds Government After AI Errors in $440K Report - News story about consulting firm refunding government after using AI for compliance analysis. · #ai-ethics #business
- B KGCreator - RDF Knowledge Graph Generation Tool - Python tool automating RDF knowledge graph generation from raw text with semantic extraction. · #knowledge-graphs #nlp #python
- B Prompt Patterns - Prompt Engineering as Design Patterns - Guide to prompt engineering using design pattern methodology for structured thinking systems. · #prompt-patterns #design-patterns #llm
- B LangChain Project Lab Book - Code Examples - Example code and implementations for integrating LLMs with external systems using LangChain. · #langchain #llm-integration #python
- B Awesome Copilot - GitHub's Collection - Community-contributed instructions and agent configurations for GitHub Copilot enhancement. · #copilot #dev-tools
- B Awesome: Master Index of Curated Topic Collections - Meta-directory of crowdsourced awesome lists covering nearly every technology and interest. · #reference #learning
- B Chroma: Open-Source Vector Database for AI - Search infrastructure and vector database purpose-built for retrieval-augmented generation. · #rag #vector-database #infrastructure
- B iroh: P2P with QUIC + NAT Traversal - Rust library adding QUIC protocol and automatic NAT traversal to peer-to-peer applications. · #infrastructure #dev-tools · M
- B Swarms - Agent Orchestration Framework - Open-source Python/Rust framework for production-ready multi-agent systems with hosted API. · #llm-agents #frameworks
- B CodeMender: AI Agent for Code Security by Google DeepMind - AI tool that automatically identifies and fixes critical security vulnerabilities in code. · #coding-agent #security
- B Free Programming Books - Massive open repository of free programming books covering languages, frameworks, science. · #learning #books · M
- B Lambda: On-Demand NVIDIA GPU Cloud - Rent H100, H200, B200 GPUs hourly for training and fine-tuning models with no egress fees. · #infrastructure #dev-tools
- B Promptify - Prompt Engineering & Versioning Platform - Tool suite for prompt engineering, versioning, and extracting structured output from LLMs. · #prompt-eng #llm-tools #versioning
- B Superpowers: Using Coding Agents in October 2025 - Practical insights on integrating AI agents into daily coding workflows and productivity. · #coding-agent #workflow
- B Chunking and Embedding Strategies in RAG - Guide to optimizing retrieval-augmented generation through smart chunking and embedding. · #rag #ai-tools
- B Awesome LLM Reasoning - From Chain-of-Thought to o1/R1 - Curated collection of LLM reasoning techniques from basic prompting to advanced models. · #reasoning #llm-research #prompt-eng
- B BitNet 1.58B Model (Quantized) on Hugging Face - Quantized LLM variant optimized for efficient local inference with llama.cpp framework. · #llm-models #llm-local
- B Depix - Recover Text from Pixelated Screenshots - Proof of concept for recovering plaintext from pixelated or redacted screenshot images. · #cybersecurity #privacy #image-analysis
- B No Science, No Startups: The Innovation Engine We're Switching Off - Steve Blank on how university science drives startup innovation and risks of defunding. · #startup-strategy #philosophy
- B auto_context - LLM Context Generation - Tool for automatically creating effective LLM prompts and contexts from text datasets. · #llm-tools #prompt-eng
- B Awesome Prompts - ChatGPT Collection - Curated list of high-performing ChatGPT prompts from GPT Store with security patterns. · #prompt-eng #collections
- B ChatGPT System Prompts Collection - Archive of extracted GPT system prompts and research on prompt injection for security. · #prompt-eng #security
- B Hacking the Humane AI Pin - Reverse engineering and hacking the Humane AI Pin after product shutdown announcement. · #ai-hardware #security · M
- B Install CockroachDB on Mac - Documentation - Official installation guide for distributed SQL database CockroachDB on macOS systems. · #infrastructure #dev-tools
- B llama.cpp Guide: Running LLMs Locally on Any Hardware - Comprehensive tutorial for setting up and running quantized LLMs on personal hardware. · #llm-local #dev-tools
- B Meta Superintelligence's First Paper Analysis - Discussion of Meta's new AI research paper and its surprising implications for models. · #ai-research #llm-models
- B Outlines - Structured LLM Outputs - Python library for constraining models to generate structured, validated JSON outputs. · #llm-libraries #structured-output
- B Tree of Thoughts - Hierarchical LLM Problem Solving - Implementation of tree-of-thoughts prompting technique improving LLM reasoning by 70%. · #prompt-eng #reasoning #llm-agents
- B Awesome ChatGPT: Curated Resources Directory - Comprehensive index of ChatGPT tools, prompts, applications, and community resources. · #llm-resources #ai-tools
- B neutts-air: On-Device Text-to-Speech by Neuphonic - Lightweight TTS model running locally on device without cloud dependencies for audio. · #ai-tools #audio
- B SelfHealingLLM - Self-Correcting Experiments - Research repository exploring self-healing and error-correction mechanisms in models. · #llm-research #error-correction
- B Simon Willison's Review of nanochat - In-depth discussion and technical review of Karpathy's full-stack LLM implementation. · #llm-implementation #ai-tools
- B ChatKit JS: OpenAI Chat UI Library - JavaScript library for building chat interfaces and applications with OpenAI models. · #coding-agent #dev-tools
- B SuperPrompt - AI Agent Prompt Engineering - Repository exploring prompt engineering to understand and improve AI agent behavior. · #llm-prompting #llm-agents #ai-research
- B Hyperflask - Full-Stack Web Framework - Integrated Python framework combining dev environment, UI, and deployment for apps. · #web-framework #python
- B OpenAI Codex: Terminal Coding Agent - Lightweight AI coding agent running in terminal for assisted development workflows. · #coding-agent #dev-tools
- B Scripting Intelligence - Book Code Examples - Companion code repository for book on Web 3.0 information gathering and processing. · #ai #web-scraping
- B Cognitive Computing Book - Code Examples - Repository of executable code samples from Mark Watson's cognitive computing book. · #ai #learning
- B maciejjankowski/SuperPrompt - AI Agent Prompt Engineering - MJ's fork of SuperPrompt for exploring AI agent behavior and prompt optimization. · #llm-prompting #llm-agents #personal-work
- B Mark Watson - Open Source AI Research - Portfolio of 20+ AI books and 55 patents by independent researcher and developer. · #ai-research #open-source #author
- B Perplexity Prompt Guide - API documentation and best practices for effective prompting with Perplexity AI. · #api-docs #prompt-eng
- B Typesense: Open-Source Search Alternative to Algolia - Fast in-memory search engine providing open alternative to Algolia and Pinecone. · #search-engine #infrastructure
- B PromptFlow - LLM App Development - Microsoft framework for building, testing, and deploying high-quality LLM apps. · #llm-frameworks #ml-ops
- B Hyperflask - GitHub Repository - Open-source repository for Hyperflask full-stack web development framework. · #web-framework #python
- B llama.cpp - Efficient Local LLM Runtime - Comparison of llama.cpp versus Ollama/LM Studio for running models locally. · #llm-local #inference
- B Spec-Driven Development for AI - Martin Fowler article on spec-driven AI patterns using Kiro and spec-kit. · #ai-dev #best-practices
- B Codex is Live in Zed - Announcement that OpenAI's Codex AI agent is integrated into Zed editor. · #code-editor #ai-agents
- C Awesome AIGC Tutorials - LLM & AI Painting Resources - Curated collection of tutorials covering language models, generative image AI, and applications. · #aigc #tutorials #generative-ai
- C GitHub Copilot Adventures: Coding Challenges - Educational coding challenges designed for learning and practicing Copilot-assisted development. · #coding-agent #learning
- C easy-invoice-pdf: Browser-Based Invoice Generator - Free open-source invoice tool with no signup, customizable templates, instant PDF export. · #dev-tools #productivity
- C Dr. TLA+ - Algorithm Learning Series - Educational series teaching algorithms and distributed system specifications using TLA+. · #formal-methods #algorithms · M
- C Prevent white screen before loading page in chromium? - Technical Q&A on browser rendering optimization and reducing initial white screen delay. · #frontend #browser-perf #chromium
- C Practical AI Development with Racket - Online book teaching artificial intelligence development using the Racket Lisp dialect. · #ai #racket
- C Lean 4 - Programming Language - Functional programming language with built-in theorem proving for formal verification. · #formal-verification #programming · M
- C pdfly: PDF Manipulation CLI Tool - Command-line tool for extracting metadata and manipulating PDF files programmatically. · #dev-tools #productivity
- C Why Racket? Why Lisp? - Essay on the philosophy and practical advantages of Lisp-family programming languages. · #lisp #programming
- C System Prompts Leaks - Extracted Prompts - Reverse-engineering research documenting system prompts from Claude, GPT, and Gemini. · #prompt-eng #reverse-engineering
- C Lean Installation Guide - Quick-start instructions for installing Lean 4 via VS Code with syntax highlighting. · #programming #lean
- C Ask HN: Abandoned Projects That Died Too Soon - Hacker News thread on promising projects that were abandoned or failed prematurely. · #reference #product-history
- C KGN Text UI - Knowledge Graph Navigator - Text-based interface for navigating and querying knowledge graphs in Common Lisp. · #knowledge-graphs #common-lisp #ai-tools
- C Mark Watson - Racket AI Book Code - Example code repository for AI and machine learning implementations in Racket. · #racket #ai-code #lisp
- C Why Your Boss Isn't Worried About AI - Essay exploring organizational perspectives on AI adoption and impact. · #ai-strategy #management
- C ChatGPT Blueprints - (unverified - PDF content not fully readable) · #prompt-eng #templates
- D Apple pico-banana-400k - (unverified - page not reachable or private Apple repository) · #ml-research #apple
- D Awesome GPT Super Prompting - Jailbreaks & Security - (unverified - page not reachable) · #llm-security #jailbreaks #adversarial · archived
- D Senior Fullstack Software Engineer – Global Automation - (unverified - page not reachable) · #recruitment #freelance #fullstack
-
D Visual Intro to Machine Learning Part 1 - (unverified - page not reachable) · #ml-learning #visualization ·
dead - F (unverified - page not reachable) - (unverified - page not reachable) · #unknown
- D CLAVIER-36 - (unverified - page not reachable) · #music-hardware · M
- B Installing LLM on Raspberry Pi Zero 2 W - Step-by-step guide to deploy local LLM on RPi Zero 2 using Ollama; OS optimization and setup instructions. · #raspberry-pi #edge-computing #tutorial · M
- B Building a Tiny Local AI Agent Desk Toy - Project guide for creating Raspberry Pi-based AI agent running fully offline on local hardware. · #ai-hardware #maker-project #electronics · M
- B MicroPythonOS: Operating System for Microcontrollers - Complete OS for ESP32 and similar microcontrollers, inspired by Android and iOS. · #embedded-systems #electronics · M
- B Reverse Engineering Amazon Kindle Web Reader - Technical breakdown of deobfuscating and removing DRM from Amazon's web Kindle. · #reverse-engineering #drm · M
- C ADS-B Massive: Real-time Flight Tracking Visualizer - ClickHouse-powered visualization of real-time aircraft, ships, weather, and geospatial data. · #data-visualization #aviation · M
- C Reddit: Using iPhone Pro LiDAR for 3D Scanning - Community discussion exploring practical applications of iPhone Pro LiDAR for 3D scanning. · #3d-scanning #hardware · M
- C Obudowa do wyświetlacza e-Paper 7.5" - Waveshare - Plastic magnetic enclosure for 7.5-inch e-ink display with USB cutout for electronics. · #electronics #hardware · M
- C shadPS4: PlayStation 4 Emulator - Cross-platform PS4 emulator written in C++ for Windows, Linux, macOS, and FreeBSD. · #emulation #gaming · M
- B When Money Is Abundant, Knowledge Is The Real Wealth - LessWrong essay on how resource bottlenecks shift from money to knowledge as scarcity changes. · #economics #philosophy
- B How to Drop the Ball Without Letting Anyone Down (Including You) - TED essay on releasing perfectionism through strategic delegation and realistic expectations. · #personal-development #productivity
- B Psychiatra: Osoby z ADHD to mistrzowie kryzysu - Polish article on ADHD as adaptive advantage: how ADHD brains excel under crisis pressure. · #adhd #psychology
- B How To Figure Out What You're Not Good At - Blog post on self-awareness and identifying genuine limitations versus limiting beliefs. · #personal-development #self-awareness
- B Tinnitus Neuromodulator - Free Therapy - Interactive audio tool using neuromodulation techniques to mask and manage tinnitus. · #audio #health
- C Aphantasia and Psychedelics - Explores how people without mental imagery experience hallucinogens; neurological perspective on perception. · #psychology #neuroscience #psychedelics
- C First Shape Found That Can't Pass Through Itself - Mathematical breakthrough solving century-old geometry puzzle about self-intersection properties. · #mathematics #geometry #science
- C 10,000 Pushups: A Silly Exercise Quest That Changed My Life - Personal narrative on completing extreme fitness challenge and discovering life transformation. · #fitness #personal-challenge
- C All-Natural Geoengineering with Frank Herbert's Dune - Essay exploring Dune's ecological engineering concepts and terraforming through living systems. · #sci-fi #worldbuilding
- C Can I Give You Some Advice? - Science article exploring psychology and philosophy of advice-giving and decision-making. · #psychology #advice #communication
- C Dark Patterns: Buying a Bahncard at Deutsche Bahn - Case study analyzing dark UI patterns used by Deutsche Bahn and Adobe in subscriptions. · #ux-design #consumer-advocacy
- C Ambi - Ambient Sounds for Sleep/Focus - Minimalist web app providing ambient soundscapes and white noise for wellness. · #audio #wellness · M
Across all years: Highlights · Index-All-Years
2025: Home · Best-Of · By tag · By category · For Maciek · Dead links · Noise
Other years: 2018 · 2019 · 2020 · 2021 · 2022 · 2023 · 2024 · 2026
Across all years: Highlights · Index-All-Years
2025: Home · Best-Of · By tag · By category · For Maciek · Dead links · Noise
Other years: 2018 · 2019 · 2020 · 2021 · 2022 · 2023 · 2024 · 2026