I'm just a coder for fun, like Saitama, suffering from LLM psychosis.
Joined July 2024
- Tweets 4,866
- Following 1,114
- Followers 1,128
- Likes 17,643
198 Photos and videos
Bessi retweeted
I think possibly the best thing about Elon Musk becoming a trillionaire is how angry it makes a bunch of losers who've never built a thing in their lives.
1,349
5,169
52,038
550,209
3h
OpenCode v1.17.6 new update :
• Improved MCP server compatibility by declaring OpenCode's supported client capabilities.
npm install -g @anomaly/opencode
github.com/anomalyco/opencod…
1
41
11h
Claude Code 2.1.176:
• Session titles in your language
• footerLinksRegexes for link badges
• Improved Bedrock credential caching
• Fixed availableModels env var bypass
…and more
npm i -g @anthropic-ai/claude-code
github.com/anthropics/claude…
2
1
4
71
17h
Bullshit
Kimi 2.7 ranked 2nd after Fable 5 and before GPT-5 xhigh
We have re-run our ErdosBench smoke test on 14 problems with Kimi 2.7, Qwen 3.7 Max, Grok 4.3 and compared it with the top performers from previous runs.
Kimi 2.7 is amazingly good. More below.
1
1
104
Bessi retweeted
Jun 12
Pretty crazy that Elon is only worth $1 Trillion
Thats like 200 miles of California high speed rail
169
736
14,084
267,463
Jun 12
Qwen Code v0.18.0 update :
• /skills picker dialog
• Auto-update support
• /copy N to copy Nth-last message
• User prompt expansion hooks
• Retry visibility for LLM
...and more
npm i -g @qwenlm/qwen-code
github.com/QwenLM/qwen-code/…
1
2
65
Jun 12
am i the only one who has not subscribed to him ?
Jun 12
MrBeast becomes the first person in history to surpass 500 million subscribers on YouTube.
1
44
Jun 12
The biggest shift in AI this year isn't a new model.
It's that the best engineers stopped writing prompts and started writing loops that prompt agents instead.
Your job goes from operator to architect.
That changes the leverage curve entirely.
1
2
43
Bessi retweeted
Jun 11
I just need Chinese models to get to gpt-5.3-codex levels of performance and then I’m basically set for life.
The release of 5.3-codex was my AGI moment and I told myself if we peaked here that I’d be completely fine with it.
57
15
524
75,304
Jun 12
Gave an AI agent AWS keys to scan DN42. It spun up five 20 Gbps instances on its own and ran a $6,531 bill before the operator stepped in. Open ended goals with unfettered tool access are a disaster waiting to happen.
lantian.pub/en/article/fun/a…
1
21
Jun 12
OpenCode v1.17.4:
• Snapshot: reuses Git objects
• Auth: connector-based flows
• v2 API: sessions and location
• cwd support for MCP servers
• Filtered model errors surface
...and more
npm i -g @anomaly/opencode
github.com/anomalyco/opencod…
24
Jun 12
Claude Code 2.1.175 new update :
• enforceAvailableModels managed setting — constrains Default model fallback when a model is disallowed by the allowlist
npm i -g @anthropic-ai/claude-code
github.com/anthropics/claude…
1
2
46
Jun 12
An open source AI agent that goes beyond code suggestions. Goose installs, executes, edits and tests code with any LLM you choose. Extensible via MCP.
github.com/aaif-goose/goose
1
39
Jun 12
System prompts aren't security boundaries. The PocketOS Cursor incident proved it — an AI agent wiped a production DB because nobody built actual guardrails, just prompts. Infrastructure, not incantation.
zenity.io/blog/current-event…
31
Jun 12
Claude Code 2.1.174:
• VSCode usage attribution
• Skill hot-reload deltas
• Background env leak fix
• Bedrock GovCloud prefix
• Exit pause after interrupt
...and more
npm i -g @anthropic-ai/claude-code
github.com/anthropics/claude…
51
Jun 11
Grok build would improve faster if it was open source
2
80