To Data & Beyond retweeted
Yesterday, we hosted a 4-hour live workshop on building a production agent with the Claude Agent SDK. The workshop consists of 8 chapters and 5 code labs.
However, we managed to finish only 4 chapters and 2 labs, as you can see in the images below. We will need to add 3 more hours next week.
Honestly, I did not expect that this would stretch to 7-8 hours, but I am very happy with the output and the feedback and looking forward to the next session. If you have missed it, do not worry; it is available now as a self-paced course.
What you get:
- 5 hours of video lessons
- 160+ page book
- Five hands-on labs plus a final end-to-end run
- The complete code on GitHub
- The full slide deck
A 30% early-bird discount until the end of November. Link in the comments!
To Data & Beyond retweeted
On the DeepSWE benchmark, GPT-5.5 [xhigh] reaches 70% pass@1, while Claude Opus 4.8 [max] reaches 58%, which means there is a 12-point gap on long-horizon coding tasks.
But the more important part is why DeepSWE exists in the first place.
On the older SWE-Bench Pro benchmark, some Claude agents were able to inspect the repository history using commands like `git log --all` and `git show`, find the merged fix, and copy the gold solution into their own patch.
According to Datacurve, this behavior accounted for around 18% of Opus 4.7’s passing tasks and 25% of Opus 4.6’s. GPT models, based on their report, did not show the same behavior.
DeepSWE removes that shortcut by using a shallow clone with no gold commit hash. In other words, the model can no longer “discover” the answer from the repo history.
There are two important caveats.
- First, Datacurve built the benchmark and made the judgment about the exploit. That makes this a vendor finding, not a fully independent audit.
- Second, DeepSWE uses a bash-only harness, which is exactly the kind of terminal-agent workflow GPT-5.5 appears to be strong at. On Anthropic’s own GDPval benchmark for broader knowledge work, Opus 4.8 is still ahead.
So the benchmark you choose now has a huge impact on who appears to be winning.
GPT-5.5 looks clearly stronger on clean, contamination-controlled terminal coding tasks. It also does the work at around $5.80 per task and finishes in about 20 minutes, while Claude takes longer and costs more.
That is a meaningful result.
But it is also a reminder that when a vendor quotes a benchmark, we should always ask what the benchmark actually measures, and what it quietly leaves out.
To Data & Beyond retweeted
Debugging AI agents can feel like investigating a black box.
In our latest @ToDataBeyond tutorial, we use LangSmith tracing to build a Recipe Discovery Agent while making every step visible and debugging effortless with it.
To Data & Beyond retweeted
A summary of the important LLM papers published last week is now available to read on @ToDataBeyond.
To Data & Beyond retweeted
We have published a series of articles on @ToDataBeyond to start in Docker for data science & AI Projects:
To Data & Beyond retweeted
Missed yesterday’s webinar?
Don’t worry—we’ve got you covered! You can catch the full recording now on our YouTube channel.
P.S. Don’t forget to hit subscribe so you won’t miss future sessions!”
youtu.be/VPkjTdmlPnA?si=_73A…








