Isolater - Feed

HN alarvfm 11d ago

1 year of LLMs writing code for me

Personal account of using AI coding agents as primary development tool for one year. Observes recent quality threshold improvements in 2025.

HN Michelangelo11 11d ago

The "Hand Farms" of India: Fueling the AI Robot Revolution with Human Motion

Human motion capture facilities in India providing training data for humanoid robot development.

HN marcobambini 11d ago

Block-Level CRDT: The Missing Piece for Collaborative AI Agent Memory

Block-Level CRDT architecture for managing distributed memory across multiple AI agents.

HN whynotmaybe 11d ago

Microsoft exec suggests AI agents will need to buy licenses,just like employees

Microsoft executive comments on potential licensing requirements for AI agents in enterprise software.

HN sapountzis 11d ago

Unslop – Filter social media slop using local-first LLM classification

Unslop: Browser extension filtering AI-generated content from social feeds using local LLM classification. Privacy-focused, no backend required.

HN JakubKontra 11d ago

Skillsmith – Write AI coding skills once, export to all AI providers

Skillsmith enables writing AI coding skills once for export across multiple AI providers and platforms.

HN Robelkidin 11d ago

OpenAI GPT worst AI GPT/model vs. Claude/MiniAAX

Comparison of OpenAI and Anthropic models.

HN nasht100 11d ago

Show HN: Wax Crate – Browse your Apple Music library like a crate of records

Music application for browsing Apple Music library as vinyl crates.

HN jedisct1 11d ago

Swival is the AI agent I wanted

CLI AI agent prioritizing data privacy and open-source alternatives to existing coding agents.

HN joozio 11d ago

Built my AI agent a custom dashboard. Then replaced it with Fizzy

Developer replaced custom AI agent dashboard with Fizzy tool.

HN latexr 11d ago

Microsoft begins removing Copilot from Notepad on Windows 11 but the AI persists

Microsoft replaces Copilot in Notepad with alternative AI writing tools on Windows 11.

HN jbatmargin 11d ago

Do MCPs improve coding agent performance?

Empirical benchmark study testing whether MCPs (Model Context Protocol) improve coding agent performance on Terminal-Bench 2.0.

HN yamafaktory 11d ago

Show HN: Formal – LLM-driven property checker, backed by Lean 4 and Mathlib

Formal: LLM-driven property checker backed by Lean 4. Identifies pure functions, generates properties, creates proofs using Mathlib.

HN camilleroux 11d ago

Show HN: Digest – a Claude Code skill for your daily dev news digest

Claude Code skill that aggregates developer RSS feeds and generates daily structured digests filtered by quality.

HN fanf2 11d ago

Ronja: Optical point-to-point data link with 1.4km range and 10Mbps full duplex

Ronja is a user-controlled optical point-to-point data link project with 1.4km range and 10Mbps duplex.

HN makaimc 11d ago

"Darwin-27B-Opus: Surpassing the Foundation Model Without Training"

Darwin-27B-Opus surpasses foundation models through evolutionary FFN breeding without additional training, achieving high GPQA scores.

HN rickcarlino 11d ago

ReBot-DevArm: open-source Robotic Arm

ReBot-DevArm is an open-source robotic arm project with full hardware and software stack for embodied AI applications.

HN PaulHoule 11d ago

Bangen is an ASCII banner renderer built on pyfiglet, rich, and Pillow

Bangen is an ASCII banner renderer built on pyfiglet, rich, and Pillow with TUI, effects, and export capabilities.

HN ianalyze 11d ago

I track what developers say about AI coding tools each week

Murmure aggregates developer sentiment from Reddit, HN, and forums into weekly intelligence reports on AI coding tools.

HN lyfeninja 11d ago

Show HN: AI Content Attestation & Verification w/ Revocable Signatures

BlkBolt technology for AI content attestation and verification using revocable signatures for agent tracking.

HN ErwanLP 11d ago

MCP Apps on Mobile: The Iframe Scroll Problem Nobody Talks About

Technical guide addressing iframe scrolling limitations in MCP Apps on mobile with design patterns for responsive UIs.

HN akavlabs 11d ago

Mitre ATLAS technique detection for LLM security in Rust

Rust implementation for detecting MITRE ATLAS techniques targeting LLM security and adversarial attacks.

HN steveharing1 11d ago

Multica: Assign issues to coding agents and track them like teammates

Multica is an open-source platform turning coding agents into managed teammates with task assignment and progress tracking.

HN appsecsanta 11d ago

The Rise of AI Pentesting Agents: A Technical Analysis (2026)

Technical analysis of AI pentesting agents evolution from PentestGPT to autonomous agents like PentAGI and XBOW.

HN aituglo 11d ago

The state of bug bounty in 2026

Essay on bug bounty trends in 2026. Discusses AI agent effectiveness for vulnerability discovery and program management challenges.

HN vanardev 11d ago

XBPP – Open standard for governing AI agent payments (Apache 2.0)

Apache 2.0 open standard for governing AI agent payment requests. Policy engine with 12 configurable checks for payment authorization.

HN atulanand94 11d ago

Open source 1040 tax software built by AI agents

Open-source tax software built and maintained by autonomous AI agents. Uses IRS publications as source, applies self-improving agent loops.

HN dev_tools_lab 11d ago

The Star Chamber: Why Multi-LLM Consensus Is Now a Necessity for Code Quality

Tool for multi-LLM code review consensus. Aggregates feedback from multiple models to identify blind spots and improve code quality assessment.

HN jonadas 11d ago

Beyond Karpathy's LLM-Wiki: The Necessity of Cognitive Governance

Essay on LLM-based knowledge management limitations. Discusses problems with AI-generated note synthesis and cognitive organization.

HN hpbyte 11d ago

Show HN: Rocky-Project Hail Mary agent skill that cut output tokens ~47%

Agent skill implementation for token compression. Reduces output tokens by ~47% while maintaining readability.

HN AkshatVirmani 11d ago

State of API Security 2026: An AI-Native Testing Perspective

Security report on 1.4M AI-driven API test executions. Maps vulnerabilities to OWASP Top 10 using agentic testing.

BL 11d ago

Enterprises power agentic workflows in Cloudflare Agent Cloud with OpenAI

Cloudflare expands access to OpenAI's frontier models via Agent Cloud platform, enabling enterprises to deploy AI agents for customer support, system updates, and report generation.

Ax Yousra Fettach, Guillaume Bied, Hannu Toivonen, Tijl De Bie 11d ago

Cards Against LLMs: Benchmarking Humor Alignment in Large Language Models

Benchmark evaluating humor alignment across frontier LLMs using Cards Against Humanity gameplay, analyzing model performance vs human baseline on comedic response selection.

Ax Zhuoyi Yang, Jiapeng Yu, Reuben Tan, Boyang Li, Huijuan Xu 11d ago

InstrAct: Towards Action-Centric Understanding in Instructional Videos

InstrAction pretraining framework for video foundation models to improve action recognition in instructional videos by addressing static bias in temporal understanding.

Ax Arda Atalik, Hui Xue, Rhodri H. Davies, Thomas A. Treibel, Daniel K. Sodickson, Michael S. Hansen, Peter Kellman 11d ago

PSIRNet: Deep Learning-based Free-breathing Rapid Acquisition Late Enhancement Imaging

Deep learning method for cardiac MRI imaging using phase-sensitive inversion recovery to reduce acquisition time and motion artifacts in late gadolinium enhancement scans.

Ax Mahdi Alizadeh 11d ago

eBandit: Kernel-Driven Reinforcement Learning for Adaptive Video Streaming

eBandit uses eBPF and multi-armed bandit reinforcement learning in Linux kernel for adaptive video bitrate selection with improved network signal visibility.

Ax Sophie Wu, Andrew Piper 11d ago

Lessons Without Borders? Evaluating Cultural Alignment of LLMs Using Multilingual Story Moral Generation

Evaluates cultural alignment of LLMs across 14 language-culture pairs using multilingual story moral generation task and dataset.

Ax Marc B\"ohlen, Sai Krishna 11d ago

Scrapyard AI

Investigates opportunities for resource-constrained AI research using obsolete yet capable discarded models from AI production cycles.

Ax Chris Hicks, Elizabeth Bates, Shae McFadden, Isaac Symes Thompson, Myles Foley, Ed Chapman, Nickolas Espinosa Dice, Ankita Samaddar, Joshua Sylvester, Himanshu Neema, Nicholas Butts, Nate Foster, Ahmad Ridley, Zoe M, Paul Jones 11d ago

Building Better Environments for Autonomous Cyber Defence

Workshop report on designing reinforcement learning environments for autonomous cyber defense applications.

Ax Fatih Cagatay Akyon, Alptekin Temizel 11d ago

SenBen: Sensitive Scene Graphs for Explainable Content Moderation

SenBen large-scale scene graph benchmark for explainable content moderation with visual grounding and sensitivity annotations.

Ax Mehran Taghian, Yunke Peng, Xing Huang, Yao Wang, Yaoyuan Wang, Wei Guo, Yuanyong Luo, Tianchi Hu, Junsong Wang, Xin Wang, Hu Liu, Yu Cheng, Ziwei Yu, Hongliang Li, Mehdi Rahimifar, Lei Yan, Xuefei Wang, Zhuang Ma, Lei Liu, Hui Yu, Anandharaju Durai Raju, Hoang Le, Hei Yi Mak, Tanzila Rahman, Shadan Golestan 11d ago

HiFloat4 Format for Language Model Pre-training on Ascend NPUs

HiFloat4 low-precision floating-point format for efficient 4-bit LLM pre-training on Ascend NPU hardware.

Ax Jinqi Luo, Jinyu Yang, Tal Neiman, Lei Fan, Bing Yin, Son Tran, Mubarak Shah, Ren\'e Vidal 11d ago

Dictionary-Aligned Concept Control for Safeguarding Multimodal LLMs

Dictionary-aligned concept control method for safeguarding multimodal LLMs against malicious queries at inference time.

Ax Cyrus Zhou, Yufei Jin, Yilin Xu, Yu-Chiang Wang, Chieh-Ju Chao, Monica S. Lam 11d ago

Scalable High-Recall Constraint-Satisfaction-Based Information Retrieval for Clinical Trials Matching

Constraint-satisfaction-based retrieval system for matching patient profiles to clinical trials with high recall and precision.

Ax Greg Nyilasy, Brock Bastian, Jennifer Overbeck, Abraham Ryan Ade Putra Hito 11d ago

AI-Induced Human Responsibility (AIHR) in AI-Human teams

Empirical study on how humans allocate responsibility in AI-human hybrid workflows using AI-assisted lending experiments.

Ax Mintong Kang, Chen Fang, Bo Li 11d ago

AudioGuard: Toward Comprehensive Audio Safety Protection Across Diverse Threat Models

AudioGuard framework for comprehensive audio safety protection including voice impersonation, speaker attributes, and compositional harms.

Ax Mohammed Maaz Sibhai, Abedalrhman Alkhateeb, Saad B. Ahmed 11d ago

MedFormer-UR: Uncertainty-Routed Transformer for Medical Image Classification

MedFormer-UR transformer with uncertainty quantification for safe medical image classification in clinical settings.

Ax Rafael da Silva, Jeff Eicher, Gregory Longo 11d ago

Temporal Dropout Risk in Learning Analytics: A Harmonized Survival Benchmark Across Dynamic and Early-Window Representations

Survival-oriented benchmark for temporal student dropout risk modeling using Open University Learning Analytics Dataset.

Ax Rafael da Silva, Jeff Eicher, Gregory Longo 11d ago