Videos

Alibaba announced the Qwen 3.8 Max Preview, a 2.4 trillion parameter open-weight model claiming performance second...

AI agent inconsistency often stems from ambiguous data near decision boundaries, not model flaws. Solutions include...

Reliability in multi-step AI agents is a control problem, not a prompting problem. Microsoft's ACE voice tutor uses...

Chinese open models like Kimiko 3 and Qwen 3.8 Max are closing the gap with closed frontier models, threatening the...

Foundation models for single-cell biology, particularly transformer-based models like SCGPT, often underperform...

Current authentication and authorization architectures designed for humans and deterministic APIs fail for...

Graphs, when properly constructed with schema and ontology, can significantly enhance AI applications by enabling...

AI-powered applications face a serious UX problem due to a wide knowledge gap between developers and users. To build...

Claude's token limits can be avoided by optimizing token consumption and model usage without increasing cost. The...

Kimi K3, a 2.8 trillion parameter open-source model from Moonshot AI, has achieved top scores on the Arena AI...

Frontier LLMs like Claude can dramatically accelerate security vulnerability discovery and patching, with Mozilla...

The Great Loops Debate at AI Engineer pits team Ian/Jeff (pro-loops, no delta) against team Dex/Greg (loops hype...

Claude Community Ambassador Dom from Australia shares his journey from attending an event to becoming an ambassador,...

Thinking Machines' Inkling model is a mid-range open-weight LLM that falls short of cutting-edge performance but...

Claude's Fable 5 design agent combined with a free skill system can build stunning, interactive websites in one...

Kun, a former elite engineer at Meta and Atlassian, describes his agentic engineering setup that uses a...

The video explains how to use Claude Cowork, a desktop app from Anthropic, to automate business tasks by setting up...

Claude Code is an agentic coding assistant that operates on a loop of observe, plan, act, and verify, using a set of...

Kimi K3 is a massive 2.8 trillion parameter open-weight model that claims to rival proprietary models like Claude...

An AI agent called Aiden, built by Weco, became the top contributor in OpenAI's Parameter Golf hiring challenge,...

Prism ML's Bonsai 27B models (ternary and 1-bit) dramatically shrink a Qwen 3.6 27B base while retaining surprising...

Lila Sciences is building AI science factories that treat the physical lab as a verifier for reinforcement learning,...

Vercel's new open-source framework Eve introduces a file-system-first approach to building production-grade AI...

Combining Fable 5 with Hermes Agent enables powerful, cost-effective AI workflows by using cheaper models for data...

Quadrable is a company building computer-using agents, and they have open-sourced Quad Driver, a background agent...

Cursor trains AI models for code generation using a recursive self-improvement loop. The outer loop gathers user...

OpenAI unveiled its first hardware device, the Codex Creator Micro keyboard, designed for dictation and thread...

Anthropic engineers automate their work by following four rules: match the bottleneck to the right solution, create...

Shipping skills without evals is risky because agents are non-deterministic and skills can degrade performance....

Cursor's forward deployed engineering (FDE) team partners with enterprise customers to co-build AI-assisted software...

OpenAI's GPT-5.6 Soul is offered at both 40-50 tokens/sec on GPUs and 750 tokens/sec on Cerebras chips (18.5x...

Codex offers advanced features like model selection (Soul, Terra, Luna), thread delegation, browser integration,...

GPT-5.6 Sol and Claude Fable 5 were compared across multiple challenging tests including 3D printing, magazine...

A new high-paying AI career as an in-house AI consultant is emerging, replacing external AI agencies. The role...

Tailscale is a private network that connects all your machines, enabling a single AI agent to control any number of...

Goda Go responds to accusations that the 'Claude Certified Architect' program is a scam, clarifying that the...

Erik Meijer argues that LLMs with tool calls are intrinsically dangerous because their agentic loop can produce...

Engram CEO Dan Biderman argues that long context windows alone cannot solve the AI memory problem; models need...

Sandboxes are essential for securely executing untrusted code from AI agents, and microVMs provide the strongest...

Fable 5 is an expensive reasoning model intended for strategic, multi-step tasks, not quick chatbot usage. By...

Anthropic launched four Claude certifications: Cloud Certified Associate Foundations, Certified Developer, Certified...

Manus AI agent can build a native Windows application from scratch, including troubleshooting and deploying it, and...

Frontier AI models are increasingly capable of discovering and exploiting software vulnerabilities, driving a 'bug...

Archon, an open-source harness builder originally designed for AI coding, can be repurposed for content creation...

Phaidra engineers solved the problem of LLMs failing to handle 500,000 sensor names in large data centers by using a...

Claude Code combined with Clay solves lead generation and cold outreach by using natural language prompts to find,...

Cactus Needle is a 26-million-parameter open-source function-calling model that runs efficiently on consumer...

AI agent hallucinations can be reduced by moving guardrails from prompts into code. Five techniques—semantic tool...

Machinecraft built a 36-agent AI system called Eira that runs their entire go-to-market without a data science team...

OpenAI's ChatGPT Sol 5.6 is a powerful new model that excels at tasks like web design, form filling, and sales...