alex zhang @a1zhang · 25 Aug 2026
“Speculative Programmatic Tool Calling” Code agents waste time waiting for the LLM to finish generating code before tool calls can start. So this paper speculatively executes predictable tool calls from partial generations, while a shadow REPL tracks dependencies safely. This https://t.co/wri2Ul7Rqd
16 424Views
84Likes
8Reposts
3Replies
3Quotes
0Bookmarks
Is that a lot?
1.72×vs this author's median9 525 views is typical
69Percentile for this authorof 13 recent posts
8.04×vs 10K–100K median2 043 views is typical
43.21%Reachviews ÷ followers
0.60%Engagement rateof viewers reacted