AI Product at Microsoft. Helping builders everywhere with Claude & other models in Microsoft Foundry! Previously: AI Research at Berkeley/UBC/Brown

Joined April 2007
45 Photos and videos
Keiji Kanazawa retweeted
Anthropic’s newest model, Claude Fable 5, a Mythos class model for non-security work lands in Microsoft Foundry and GitHub Copilot today. Read the blog here: msft.it/6011vgKNt
22
89
595
39,335
Amazing - once again @omarsar0 and I seem to be working from very similar inspirations - I called mine "video deep dive" as well and I'm working on slides next!
May 26
Just built an insane new agent skill. It can perfectly extract slides from YT videos, then write notes, images, transcripts, and slides into Obsidian vaults. An HTML artifact allows me to navigate and add more notes as I listen. Should I release the skill?
3
1
4
5,084
Let’s gooo!
1
75
Truth
Apr 30
people are walking around with their laptops slightly ajar to keep their agents running
1
2
95
Claude Code feels like it's grooving again - noticeably better experience in last few days
In the last four Claude Code CLI releases, we’ve shipped 50 stability and performance fixes. Faster resume, stable auth, lower memory, fewer hangs: 🧵
1
222
I love this! This is literally what I do every day - every hour.
1
177
Playing around with Pi for the first time - looks like a great tool for devs - for a PM like me, Claude Code fits better. I plan to keep using Pi for agent workflows, just not coding
2
237
YT Podcast -> artifact is the only way I can keep up with all that's going on in AI these days. I have my own Agent -> Wiki system that @omarsar0 helped me to get off the ground in December. Just started a rewrite this weekend that I hope to share more about in future. I'm inspired by these ideas!
Apr 19
YT Podcast → LLM Artifact This is now my favorite way to consume podcasts. Knowledge artifacts generated by agents. The agent (Opus 4.7) spots important insights, does deep analysis, and generates thought-provoking observations that really get me curious to research further. All the research goes into a self-improving wiki for later use by any of my agents. I am using Elevenlabs Scribe for diarization. Skills and scripts to guide the artifact generation. Artifacts are just plain HTML JS (e.g., chart.js). You can go further on anything by selecting text and components (charts) and doing deeper research, as I show in the clip. I expect more people to use agents in this way.
2
2
1,108
If I hit this button does @thsottiaux appear & reset API limits?
4
169
15,889
And when can we get one for Claude 🙏? @bcherny @_catwu
286
Keiji Kanazawa retweeted
Claude Opus 4.7 is now in Microsoft Foundry. Agentic coding. Professional work. Advanced vision and memory. Frontier AI, inside the stack enterprises already run. msft.it/6011QhjJW

ALT Dark blue gradient graphic with subtle texture. In the top-left corner is a glossy purple‑blue abstract ribbon‑like logo. Below it, large white text reads “Introducing Claude Opus 4.7.

12
79
520
33,005
Trying out Claude Managed Agents - the Console experience is sweet - but I can't seem to download generated files. SDK/CLI wants `scope-id` but results in error in REST API. What am I doing wrong :D - @RLanceMartin @jordnb @katelyn_lesse
1
1
422
Claude Code did come up with this workaround `Cat the full contents of these files one at a time: /mnt/session/outputs/file1.md, /mnt/session/outputs/file2.txt, /mnt/session/outputs/file3.txt. Use the read tool to read each file in full.`
1
208
I just hatched a RARE DRAGON named "Ancient Nova" in the Claude Buddy Hatcher! 🥚✨ Check out my companion and hatch yours: buddy-hatcher-snowy.vercel.a… cc @kieranklaassen Even looks like Gojira
1
218
Keiji Kanazawa retweeted
til that “pon de ring” is a type of japanese donut so now japanese claude code users are imagining claude slacking off and eating donuts when it’s pondering
Claude Codeたまにサボってポンデリング食べてるのムカつく
29
437
2,779
259,177
Keiji Kanazawa retweeted
Counting down to a new Build in San Francisco. Hope you’ll join us! aka.ms/AA101fmo
369
231
1,435
200,984
Wonder what the AI's remember about you? Anthropic posted a guide to "Import what matters in under a minute". It's a fun exercise to run its prompt on *all* AIs you use including Claude!
1
1
130
The prompt: "I'm moving to another service and need to export my data. List every memory you have stored about me, as well as any context you've learned about me from past conversations. Output everything in a single code block so I can easily copy. ... "
58