10 entries
Keyword: kun-chenX post:Kun Chen
Kun Chen highlights Clapper, an open-source tool built by Calvin Grunewald. The skill allows coding agents to generate animated presentational videos by writing React code, complete with automated music composition and narration.
235bookmarks19.7kviews
Article:Kun Chen
Software engineer Kun Chen details his personal configuration for the Pi agent harness, focusing on streamlined terminal output, prompt queuing, OpenAI context compaction, and strict API cost boundaries.
From X post: Minimal Pi Agent Configuration for AI-Assisted Coding
X post:Kun Chen
Kun Chen outlines the architectural combination of Herdr and Firstmate for multi-agent workflows. Herdr manages agent lifecycles and organization, while Firstmate acts as the primary orchestrator and single point of contact.
1kbookmarks74.6kviews
Article:Kun Chen
Kun Chen argues that project-level memory files like AGENTS.md or CLAUDE.md should be treated as neural network weights. Rather than manually appending rules based on anecdotes, he applies a gradient-descent loop driven by actual session transcripts, implemented in the open-source CLI tool backpass.
From X post: Treating AGENTS.md like a neural net: Kun Chen's guide to agent memory files
X post:Kun Chen
Kun Chen recommends against immediately rewriting CLAUDE.md or AGENTS.md when switching to newly released AI models. Instead, teams should keep their existing setup initially and use the open-source CLI tool backpass to propose evidence-backed refinements derived from session transcripts.
633bookmarks57.2kviews
X post:Kun Chen
Kun Chen advocates systematically optimising AI assistant memory and skill files using a 'backward pass'. Instead of reacting to one-off anecdotes, large samples of session transcripts should be analysed to distill losses and update skills systematically.
378bookmarks9.5kviews
X post:Kun Chen
Kun Chen proposes an engineering framework for managing AGENTS.md and CLAUDE.md files: user-level preferences should be handwritten, while repository-level instructions should be treated like a neural network with a strict token budget and refined via backward passes on real session transcripts.
2.8kbookmarks93.5kviews
X post:Kun Chen
Kun Chen argues that recent frontier LLMs have deteriorated in everyday conversational quality. Newer models appear more robotic, rely heavily on jargon, produce overly verbose responses, and perform tasks that were never requested.
1.9kbookmarks292.6kviews
X post:Kun Chen
Kun Chen published his practical configuration and dotfiles for Pi, a provider-neutral agent harness for LLMs. The guide covers terminal preferences, prompt queueing, key extensions, and API cost caps for OpenAI models.
1.3kbookmarks39.5kviews
VideoVideo:David Ondrej
In conversation with David Ondrej, former Meta, Microsoft, and Atlassian engineer Kun Chen demonstrates his fully agentic development workflow. Rather than manually juggling dozens of agent sessions, he uses his custom tool FirstMate as a primary coordinator. FirstMate delegates sub-tasks to specialised crewmate agents inside Herder, an agent-aware terminal multiplexer running in WezTerm.
10 of 10 shown
Do you have an interesting article, a video, or a tweet that belongs in our knowledge base? Write to us at office@webconsulting.at.
Parts of this content were created with the assistance of AI.
