Issue #4 ·
Five things from the last few days worth your attention if you build with AI. No fluff.
Five things worth your attention if you build with AI, about 3 minutes. Join free and get 13 ready-to-paste Starter Prompts for coding agents, vibe coding and game dev right away.
MODEL
It's a 2.8T-parameter mixture-of-experts model (104B active) with a 1M-token context window. On Terminal-Bench 4.0 it scores 65.7%, just behind Claude Opus 5.5 at 66.4% and ahead of GPT-6 Astra at 58.2%.
Why builders care: It's live now in the Kimi API and in Kimi Code (pick it with /model), so you can test it on your agent tasks before the weights land.
CODING
The Nanbeige team released a dense 33B model under Apache 2.0 with a 512K context window. On its own comparison it scores 65.5 on DeepSWE against 42.5 for Qwen3.8-27B, though Qwen edges it on SWE-Bench Pro (61.7 vs 60.3).
Why builders care: For agent work, the team recommends preserve_thinking=True and XML tool calls. The BF16 weights are about 66 GB, so most laptops will want a quantized build.
AGENTS
Describe a task in plain language and the YC startup's open-source tool (MIT) builds a reusable API or MCP integration for that site, even behind a login and two-factor auth.
Why builders care: A repeat chore on a site with no API becomes a fast, predictable tool call instead of a browser agent clicking through it every run.
GAME DEV
Steam's content survey asks developers to describe generative AI in content players see, like art, sound, writing and store assets, and shows it on the store page. Epic's CEO wrote that forcing developers to write "a political statement to be used against them by the hate mob" is abhorrent.
Why builders care: Valve says AI tools used only for efficiency aren't the focus. Generated art, voice or text that ships is what you disclose.
REPO
A reference kit (Apache 2.0) for Anthropic's new advice: fold subagents into your main loop one at a time, turning their prompts into skills and their API calls into MCP tools, and compare evals after each step. It ships compare and coverage scripts plus a support-desk example.
Why builders care: One customer saw runtime roughly halve with a single loop. Keep a subagent only for context bloat, parallel throwaway work, or conflicting instructions.
That's the issue. Know a builder who'd use this? Share it on X or forward them this page. New here? Subscribe free for tomorrow's.
Five things worth your attention in AI, agents, vibe coding and game dev, every weekday. Plus 13 free Starter Prompts and the AI Builder Starter Stack PDF when you join.
Want the bigger picture? Read our guides to AI coding agents, vibe coding tools, AI game dev tools, agent frameworks, MCP servers and web scraping tools, or all guides.
