Archive - TechTalks

How Codev brings discipline to AI software development

Casual AI prompting breaks down as codebases grow. Codev introduces strict protocols and multi-model reviews to help teams ship maintainable software.

3 hrs ago • Ben Dickson

Scaling the harness: The next major bottleneck in agentic AI

Scaling LLMs hits limits when dealing with agentic AI tasks. For that, we need to look at the harness and the system built around the model(s).

Jun 2 • Ben Dickson

May 2026

What makes Cursor's Composer 2.5 a good coding model (and what are the caveats)

A deep look at the self-distillation techniques that make Composer 2.5 such a great coding model (and the hidden tradeoffs they introduce to AI…

May 26 • Ben Dickson

How prompt injection broke Nvidia's sandboxed OpenClaw agent

Research into Nvidia’s NemoClaw reveals that sandboxes don't stop AI agents like OpenClaw from leaking data. We need to rethink security from first…

May 19 • Ben Dickson

Unpacking Gemma 4’s multi-token prediction (and why you should care)

How Gemma 4’s multi-token prediction and community-driven DFlash are speeding up local LLM throughput by 3-6x.

May 12 • Ben Dickson

How to scale LLMs to 100 million tokens without blowing up memory costs

Memory Sparse Attention (MSA) scales LLM context windows to an unprecedented 100 million tokens while preserving accuracy.

May 5 • Ben Dickson

April 2026

The silent threat of Claude Code (and other AI coding assistants)

A new study reveals how AI coding assistants like Claude Code are quietly hoarding and publishing sensitive API keys to code repositories.

Apr 28 • Ben Dickson

The 'by design' security flaw of Model Context Protocol (MCP)

Security researchers have uncovered a massive architectural flaw in Anthropic's Model Context Protocol, exposing millions of AI applications to remote…

Apr 21 • Ben Dickson

The hidden trap of LLMs self-distillation

Optimizing LLMs for concise answers can destroy their ability to explore alternative solutions on difficult problems. New study reveals the hidden cost…

Apr 14 • Ben Dickson

The art of AI harness engineering

The recent leak of Anthropic's Claude Code reveals a hard truth: as LLMs become commoditized, the sophisticated engineering harness built around them is…

Apr 7 • Ben Dickson

March 2026

How GhostClaw exploits macOS and OpenClaw to steal developer credentials

As developers rush to run local AI agents on Mac Minis, GhostClaw malware exploits macOS binaries to silently harvest credentials.

Mar 31 • Ben Dickson

Inside V-JEPA 2.1, the huge upgrade to Meta's world model

AI models have historically struggled to balance motion tracking with spatial detail. Meta’s V-JEPA 2.1 solves this, pushing the boundaries of video…

Mar 25 • Ben Dickson