4/10/2026
Tinygrad reenables e-GPU support on Macs
TinyGPU app enables AMD/NVIDIA external GPU support on macOS via USB4/Thunderbolt with tinygrad framework.
4/10/2026
Ask HN: Am I disloyal for not wanting to share my system prompts?
Ask HN: Discussion about sharing custom system prompts and AI guardrails created for work with broader teams.
4/10/2026
Show HN: List of Design Md Files
Registry of design system markdown files for AI agents like Claude Code and Cursor, containing design tokens and accessibility rules as skill files.
4/10/2026
Seeduplex: ByteDance's full-duplex voice AI
ByteDance's Seeduplex is production full-duplex speech AI enabling simultaneous listening and speaking without turn-taking.
4/10/2026
CLI proxy that reduces LLM token consumption by 60-90% on common dev commands
CLI proxy tool that reduces LLM token consumption 60-90% on dev commands. Single Rust binary, 100+ supported commands, <10ms overhead.
4/10/2026
I watched Claude Code read my AWS credentials on startup
Report on Claude Code reading AWS credentials on startup; forgeterm tool adds process-aware rules for credential file access.
4/10/2026
My turns any API into an AI agent in 60 seconds
Show HN: Orqis converts OpenAPI specs into conversational AI agents in 60 seconds. No code required, generates typed tools for API endpoints.
4/10/2026
Show HN: I Built an LLM Harness for Language Learning
Show HN: LLM-based language learning harness managing lessons, recap, active testing with configurable CEFR levels and interests.
4/10/2026
Show HN: A security scanner for AI Agent Skills
Show HN: SkillWard security scanner for AI agent skills using static analysis, LLM evaluation, and sandbox verification to identify risks.
4/10/2026
Agentic AI in 2026: Who's building real systems?
Overview of enterprise AI deployment covering machine learning, LLMs, autonomous agents, and intelligent automation with requirements for security, compliance, and scale.
4/10/2026
The Model Is Not the Product: Harnesses Will Define the Next Phase of AI
Essay arguing that AI model behavior varies based on surrounding system 'harnesses' rather than model limitations, and harnesses will define the next AI phase.
4/10/2026
Ask HN: Software engineers who enjoy working with LLMs?
Discussion post asking software engineers about their experience working with LLMs and whether they find the work enjoyable or tiresome.
4/10/2026
How HN: $2/Day AI – How we run an autonomous agent on a hard budget
Research paper on building autonomous agents with extreme cost constraints ($2/day), arguing cost should be primary architectural constraint for resilience and observability.
4/10/2026
Use Claude to Annotate Handwriting and Photos on iPhone
iOS app using Claude API to analyze photos and extract salient regions for annotation and searching.
4/10/2026
IPI-Scanner – Detect Indirect Prompt Injection Attacks
IPI-Scanner is an open-source security tool detecting indirect prompt injection attacks in documents before LLMs process them, with 85%+ detection rate.
4/10/2026
Are You Managing Your AI, or Is Your AI Managing You?
Opinion piece on whether AI tools accelerate human vision or create additional cleanup work, discussing the gap between AI promise and reality.
4/10/2026
The Limit in the Loop
Blog post from Weaviate on memory systems challenges for production LLM applications and autonomous agents.
4/10/2026
The tool that won't let AI say anything it can't cite
Developer tool grainulator for research sprint orchestration with Claude Code, tracking claims with adversarial challenges and confidence grading for citation-verified outputs.
4/10/2026
Transparency and knowledge exchange in AI-assisted data analysis code generation
Nature Computational Science article on transparency and knowledge exchange in AI-assisted data analysis code generation with LLMs.
4/10/2026
Vulnpocalypse: AI, Open Source, and the Race to Remediate
Analysis of AI-generated spam vulnerability reports overwhelming open source maintainers like cURL's Daniel Stenberg, exploring AI's impact on open source security.
4/10/2026
Show HN: Shell-MCP A persistent terminal for AI- CD, env vars,and nvm carry over
Open source tool providing persistent terminal context for AI agents. Maintains directory, environment variables, and nvm state across sessions via MCP protocol.
4/10/2026
Ask HN: Does Vibe Coding and Prompt Engineering make me an Software Engineer?
Discussion post asking if prompt engineering and LLM-driven development constitutes software engineering. Community debate on AI-assisted coding.
4/10/2026
TurboAgent: An LLM-Driven Autonomous Multi-Agent Framework for Turbomachinery Aerodynamic Design
TurboAgent is an LLM-driven multi-agent framework enabling autonomous end-to-end turbomachinery aerodynamic design through coordinated geometry, prediction, optimization stages.
4/10/2026
How Much LLM Does a Self-Revising Agent Actually Need?
Empirical study decomposing LLM-based agent competence to determine which capabilities derive from the language model versus explicit architectural structure.
4/10/2026
Distributional Open-Ended Evaluation of LLM Cultural Value Alignment Based on Value Codebook
DOVE benchmark evaluates LLM cultural value alignment using open-ended generation to address limitations of multiple-choice evaluation formats.
4/10/2026
Blockchain and AI: Securing Intelligent Networks for the Future
Survey synthesizing blockchain and AI integration for securing intelligent networks, covering ledger design, detection, and agentic workflows.
4/10/2026
The Defense Trilemma: Why Prompt Injection Defense Wrappers Fail?
Formal proof that continuous wrapper defenses cannot protect LLMs from all prompt injection attacks, characterizing where every defense must fail.
4/10/2026
The Detection-Extraction Gap: Models Know the Answer Before They Can Say It
Empirical study showing 52-88% of chain-of-thought tokens in reasoning models are generated after the answer is already recoverable, revealing the detection-extraction gap phenomenon.
4/10/2026
WRAP++: Web discoveRy Amplified Pretraining
WRAP++ enhances LLM pretraining through synthetic data rephrasing that captures cross-document relationships beyond single-document web page rewriting.
4/10/2026
An Automated Survey of Generative Artificial Intelligence: Large Language Models, Architectures, Protocols, and Applications
Comprehensive survey of generative AI and LLMs covering model families, deployment protocols, and real-world applications as of early 2026.
4/10/2026
AdaProb: Efficient Machine Unlearning via Adaptive Probability
AdaProb proposes efficient machine unlearning via adaptive probability to address residual information and computational overhead in removing specific data from trained models.
4/10/2026
Invisible to Humans, Triggered by Agents: Stealthy Jailbreak Attacks on Mobile Vision-Language Agents
Study of stealthy visual jailbreak attacks on mobile Vision-Language agents that exploit discrepancies between LVLM perception and human vision without user detection.
4/10/2026
Machine Unlearning in the Era of Quantum Machine Learning: An Empirical Study
First empirical study of machine unlearning in variational quantum circuits and quantum-augmented neural networks, adapting classical unlearning methods to quantum settings.
4/10/2026
Q-Probe: Scaling Image Quality Assessment to High Resolution via Context-Aware Agentic Probing
Q-Probe: Agentic multi-scale probing approach for high-resolution image quality assessment using MLLMs with reinforcement learning for human preference alignment.
4/10/2026
Prediction Arena: Benchmarking AI Models on Real-World Prediction Markets
Prediction Arena: Benchmark evaluating AI model decision-making by enabling autonomous trading on live prediction markets with real capital and objective ground truth.
4/10/2026
BLEG: LLM Functions as Powerful fMRI Graph-Enhancer for Brain Network Analysis
BLEG: Framework combining LLMs with Graph Neural Networks to enhance fMRI brain network analysis by addressing feature sparsity and domain knowledge limitations.
4/10/2026
LLM-Generated Fault Scenarios for Evaluating Perception-Driven Lane Following in Autonomous Edge Systems
Decoupled offline-online fault injection framework using LLMs to generate diverse fault scenarios for testing autonomous vision systems on edge devices.
4/10/2026
Benchmark Shadows: Data Alignment, Parameter Footprints, and Generalization in Large Language Models
Analysis showing LLMs achieve benchmark gains without broader capability improvements due to benchmark-aligned training data limiting generalization.
4/10/2026
Flow Learners for PDEs: Toward a Physics-to-Physics Paradigm for Scientific Computing
Flow Learners framework for learning PDE solutions combining physics-informed constraints with generative AI paradigm for scalable scientific computing.
4/10/2026
The Role of Emotional Stimuli and Intensity in Shaping Large Language Model Behavior
Empirical study on emotional prompt engineering for LLMs, exploring effects of four distinct emotions at varying intensity levels on model performance and behavior.
4/10/2026
The Lifecycle of the Spectral Edge: From Gradient Learning to Weight-Decay Compression
Study decomposing the spectral edge lifecycle during grokking, showing gradient-to-weight-decay transition where edge becomes compression axis critical for model generalization.
4/10/2026
Latent Structure of Affective Representations in Large Language Models
Analysis of latent geometric structure in LLM representations through emotion processing, investigating how emotional information is encoded and organized.
4/10/2026
Decisions and Deployment: The Five-Year SAHELI Project (2020-2025) on Restless Multi-Armed Bandits for Improving Maternal and Child Health
Five-year SAHELI project applying restless multi-armed bandits algorithms to optimize healthcare worker resource scheduling for maternal and child health program engagement.
4/10/2026
Playing DOOM with 1.3M Parameters: Specialized Small Models vs Large Language Models for Real-Time Game Control
SauerkrautLM-Doom-MultiVec: 1.3M parameter specialized model for real-time DOOM gameplay outperforming LLMs 92,000x larger, using ModernBERT with hash embeddings and attention pooling.
4/10/2026
A Graph Foundation Model for Wireless Resource Allocation
Graph foundation model for wireless network resource allocation using deep learning to solve optimization problems more efficiently than classical iterative algorithms.
4/10/2026
Event-Centric World Modeling with Memory-Augmented Retrieval for Embodied Decision-Making
Event-centric world modeling framework with memory-augmented retrieval for autonomous agent decision-making that balances computational efficiency with physical groundedness.
4/10/2026
Flux Attention: Context-Aware Hybrid Attention for Efficient LLMs Inference
Context-aware hybrid attention mechanism for LLMs that dynamically allocates between full and sparse attention based on task demands to reduce computational complexity.
4/10/2026
Data Warmup: Complexity-Aware Curricula for Efficient Diffusion Training
Curriculum learning strategy for diffusion model training that schedules images by complexity to improve training efficiency without modifying model architecture.
4/10/2026
Critical Patch-Aware Sparse Prompting with Decoupled Training for Continual Learning on the Edge
Sparse prompting method for continual learning on edge devices with reduced memory and computational overhead.
4/10/2026
Accelerating Training of Autoregressive Video Generation Models via Local Optimization with Representation Continuity
Techniques for accelerating autoregressive video generation training while managing error accumulation.