6d ago
Reports of RAG's death have been greatly exaggerated
Technical post discussing RAG vs fine-tuning approaches for LLMs, referencing Andrej Karpathy's markdown wiki idea and introducing Atomic tool.
6d ago
Show HN: LLM inference slowdown fixed (177 experiments, +37% attention) – in 48h
Claim of 37% attention improvement in LLM inference optimization via 177 experiments. Title only, no technical details.
6d ago
Elevated Errors on Claude.ai, API, Claude Code
Status update: Claude.ai and API experiencing elevated errors. Operational status report, no technical insights.
6d ago
Mapping GPUs to LLMs (and back): A bandwidth-based estimator for local inference
Technical guide for selecting GPUs and LLM models for local inference, addressing hardware-model compatibility for cost-effective local deployment.
6d ago
Show HN: IDE like, local API client
Postchi is an IDE-like local API client built with Tauri and React for focused API development.
6d ago
Agentic Engineering Methodology – Structured AI-Assisted Dev (Karpathy, Osmani)
Methodology by Karpathy and Osmani for structured AI-assisted development using agentic approaches. Title only, no details.
6d ago
KillBench shows every major LLM is biased against Americans
Claims major LLMs show bias against Americans. Title only, no substantive content provided.
6d ago
Sony Is Removing Many Popular Features from Its Free OTA TV Options
Sony removing TV Guide and related features from older BRAVIA OTA models starting May 2026.
6d ago
How to hit your Claude weekly limit so you can go outside and touch grass
Humorous personal essay about obsessive Claude usage and taking a break. No technical content or insights.
6d ago
Show HN: ClaudeWatch – macOS menu bar usage monitor for Claude with notch pet
ClaudeWatch is macOS menu bar tool for tracking Claude Code token usage, rate limits, and costs with notch pet.
6d ago
Show HN: Voiden – API Workflows as Markdown Files (Obsidian × Curl)
Voiden is an open-source offline API tool that executes API requests as Markdown files with Git versioning, combining Obsidian-style workflows with curl functionality.
6d ago
ModelCascade – Route LLM calls to your own GPU first, cloud second
ModelCascade is open-source router for LLM calls that handles 74% of requests locally on GPU, escalating to cloud API when needed.
6d ago
Show HN: SafeWeave – 8 security scanners as an MCP server inside your AI editor
SafeWeave is an MCP server integrating 8 security scanners into AI editor for code analysis.
6d ago
LLM pricing is 100x Harder than you think
Discussion of LLM API pricing complexity and hidden variables. Title only, no substantive content.
6d ago
Project Think: building the next generation of AI agents on Cloudflare
Cloudflare Project Think: framework for building next-generation AI agents on their platform. Title only, no technical details.
6d ago
Communicating Our Research with Stakeholders to Achieve Alignment and Trust
Framework for communicating research to stakeholders in smart cities using structured, evidence-driven approaches.
6d ago
GPT-5.4 pro solves erdos problem #1196
GPT-5.4 Pro claims solution to Erdos problem #1196 with short proof; community verification requested.
6d ago
Friday, self-evolving assistant, only CC $100 plan, no agent framework
Friday is a self-evolving AI assistant running 24/7 on personal machines via Claude Code CLI and Telegram, learning autonomously.
6d ago
Show HN: A book that builds GPT-2, Llama 3, DeepSeek from scratch in PyTorch
Book implementing GPT-2, Llama 3, and DeepSeek architectures from scratch in PyTorch with progressive examples and real weight loading.
6d ago
I ran 5 social engineering attacks on AI. The failure modes are human
Analysis of LLM jailbreak attempts as social engineering rather than code exploits, examining failure modes through psychological manipulation lens.
6d ago
Vibe Coding Fails
User documents patterns in maintaining Claude-based agent loops with approval workflows and identifies failure modes in vibe coding.
6d ago
Playmakerly – AI agent that runs football prediction leagues in Slack
AI agent managing football prediction leagues within Slack. Demonstrates practical agent application for sports forecasting.
6d ago
Show HN: A /speak slash command for Claude Code that reads replies aloud
Claude Code /speak command that reads AI assistant replies aloud using system text-to-speech on macOS.
6d ago
Why the World Needs Open Source AI
Opinion on necessity of open source AI development. Advocacy piece for open AI model accessibility.
6d ago
Are the Costs of AI Agents Also Rising Exponentially? (2025)
Analysis of whether AI agent operational costs are rising exponentially. Cost economics for deployed AI agents.
6d ago
Show HN: RAG chunking playground: visualize how your docs get split
Interactive tool for visualizing RAG document chunking strategies. Developer tool for optimizing retrieval-augmented generation systems.
6d ago
Allbirds is turning into an AI compute provider
Brief headline about Allbirds pivot to AI compute provider. Insufficient content for evaluation.
6d ago
Xata: Open-source Postgres platform with copy-on-write branching
Xata is an open-source, cloud-native Postgres platform for self-hosting on Kubernetes with branching capabilities.
6d ago
Bosses say AI boosts productivity – workers say they're drowning in 'workslop'
Opinion piece on AI-generated work quality issues in workplaces. Workers report flawed AI outputs requiring heavy corrections.
6d ago
Allbirds announces pivot from shoes to AI, stock explodes 175%
Allbirds stock jumps on announcement of shift to AI compute business with $50M funding target.
6d ago
Making prompts longer did not help. Making the task contract explicit did
Benchmark study on Python code tasks shows explicit task contracts improve LLM performance better than longer prompts alone.
6d ago
Show HN: LLM-primer – pre-warmed Claude Code session pool, zero startup wait
LLM-primer: pre-warmed Claude Code session pool eliminating 30-60s startup latency. Developer tool for maintaining persistent agent contexts.
6d ago
MCP as Observability Interface: Connecting AI Agents to Kernel Tracepoints
Using Model Context Protocol as observability interface to connect AI agents to kernel tracepoints. Technical approach for agent monitoring and debugging.
6d ago
Ask HN: Why/how do you use LLMs to draft tech blogs?
Survey asking how developers use LLMs to draft technical blogs. Community research on LLM content creation practices.
6d ago
Is "build an audience before a product" bad advice? 3 AI models debated it
Three AI models debate whether building an audience before a product is bad advice. Social commentary without technical depth.
6d ago
1-week challenge using resilient LLMs (starts this Sunday)
ResilientLLM library for production-ready LLM integration handling failures, rate limits across multiple providers. Developer tool for reliable agent and LLM applications.
6d ago
Show HN: Lazyagent – TUI for to watch all your AI coding agents
Lazyagent is a terminal UI tool for monitoring multiple AI coding agents (Claude Code, Codex, OpenCode), displaying their events and tool calls in a unified interface organized by working directory.
6d ago
Show HN: Chat-rs, yet another LLM provider
Chat-rs: Rust LLM inference provider with streaming, tool calling, model routing, and human-in-the-loop. Open source developer tool with agent support.
6d ago
Show HN: Agent first social media scheduling
Agent-first social media scheduling tool. Application of AI agents to content management automation.
6d ago
AI ruling prompts warnings from US lawyers: Your chats could be used against you
Legal warning that AI chats could be used as evidence in court. Policy/legal news without technical content.
6d ago
AI learns language from skewed sources. That could change how we speak and think
Analysis of how LLM training on written text skews language representation compared to unscripted conversation. Research on language model training data bias.
6d ago
Show HN: Monogate – EML operator family, hybrid framework, 108-node sin(x)
36-hour implementation extending arXiv:2603.21852 symbolic regression paper. Creates hybrid EML operators achieving 52-74% node reduction with machine-precision results.
6d ago
Open-source AI cost router – cut API bills 30-50% with automatic failover
CLI tool for routing AI tasks across multiple providers with persistent memory, reducing API costs 30-50% with automatic failover.
6d ago
Show HN: Pseudonymizing sensitive data for LLMs without losing context
Context-preserving pseudonymization proxy for Claude to mask sensitive data in security analysis workflows.
6d ago
Study: Back-to-basics approach can match or outperform AI in language analysis
Study showing grammar-based approach (LambdaG) matches AI models in authorship detection with greater transparency and lower cost.
6d ago
Allbirds, Inc. Announces Expansion into AI Compute Infrastructure
Allbirds footwear company announces AI compute infrastructure investment after selling brand assets.
6d ago
Ask HN: Is Claude Getting Worse?
User discussion reporting perceived quality degradation in Claude model performance and token limits.
6d ago
Bws-MCP-server – giving AI agents a controlled path to Bitwarden secrets
MCP server providing AI agents controlled access to Bitwarden secrets using ephemeral credential management.
6d ago
Trysil – A lightweight ORM for Delphi with soft delete and change tracking
Object-relational mapping library for Delphi with soft delete and change tracking features.
6d ago
Show HN: WorkProtocol – Where AI agents do real work for real pay
Infrastructure platform for autonomous AI agents to claim jobs, deliver work, and receive payment in USDC with escrow verification.