Tootfinder

Reducto, which uses OCR with vision language models to convert complex documents into inputs for LLMs, raised a $75M Series B led by a16z at a $600M valuation (Stephanie Palazzolo/The Information)
https://www.theinformation.com/articles/startup-using-ai-tran…

The Startup Using AI to Translate Documents Into Data
If you’ve ever uploaded a picture of a receipt to an expense report or read a PDF of a book online, you’ve likely used optical character recognition, a decades-old technique that converts images of typed, handwritten or printed text into text that’s editable on a computer.OCR might not sound ...

Any advice / opinions on wiki hosting providers?
Doesn’t have to be free, but user-friendliness is key: it’s for my dad, who is computer-competent but not deeply technical. Friendly editor is more important than power features. Ability to export a major plus.

SP-MoE: Speculative Decoding and Prefetching for Accelerating MoE-based Model Inference
Liangkun Chen, Zijian Wen, Tian Wu, Xiaoxi Zhang, Chuan Wu
https://arxiv.org/abs/2510.10302

SP-MoE: Speculative Decoding and Prefetching for Accelerating MoE-based Model Inference
The Mixture-of-Experts (MoE) architecture has been widely adopted in large language models (LLMs) to reduce computation cost through model sparsity. Employing speculative decoding (SD) can further accelerate MoE inference by drafting multiple tokens per step and verifying them in parallel. However, combining MoE with SD inflates GPU memory and aggravates CPU-GPU bandwidth contention during multi-token verification. Existing MoE offloading systems are SD-agnostic and do not address this bottlene…

Day 19 (a bit late): Alice Oseman
As I said I've got 14 authors to fit into two days. Probably just going to extend to 30? But Oseman gets this spot as an absolute legend of queer fiction in both novel & graphic novel form, and an excellent example of the many truths queer writers have to share with non-queer people that can make everyone's lives better. Her writing is very kind, despite in many instances dealing with some dark stuff.
I started out on Heartstopper, which is just so lovely and fun to read, and then made my way through several of her novels. The one I'll highlight here which I think it's her greatest triumph is "Loveless", which is semi-autobiographical and was at least my first (but no longer only) experience with the "platonic romance" sub-genre. It not only helped me work through some crufty internal doubts about aro/ace identities that I'd never really examined, but in the process helped improve my understanding of friendship, period. Heck, it's probably a nice novel for anyone questioning any sort of identity or dealing with loneliness, and it's just super-enjoyable as a story regardless of the philosophical value.
To cheat a bit more here on my author count, I recently read "Dear Wendy" by Ann Zhao, which shouts out "Loveless" and offers a more expository exploration of aro/ace identities, but "Loveless" is a book with more heart and better writing overall, including the neat plotting and great pacing. I think there are also parallels with Becky Albertalli's work, though I think I like Oseman slightly more. Certainly both excel at writing queer romance (and romance-adjacent) stuff with happy endings (#OwnVoices wins again with all three authors).
In any case, Oseman is excellent and if you're not up for reading a novel, Heartstopper is a graphic novel series that's easy to jump into and very kind to its adorable main characters.
I think I've now decided to continue to 30, which is a relief, so I'm tagging this (and the next post that rounds out 20) two ways.
#20AuthorsNoMen
#30AuthorsNoMen

Series B, Episode 10 - Voice from the Past
AVON: I'm sorry to have to inform you that he is himself all right.
[End of episode. Roll credits.]
https://blake.torpidity.net/m/210/623 B7B3

I have to say, spending a decade organizing (for bike/ped/transit infrastructure) has done more for my understanding of how the world _actually_ functions than working at any company or startup, taking any classes*, or really anything else. It's the weird interactions between businesses, the public, technology, and the political landscape that really shape our society.
* Not that classes aren't useful! I took a law class in college and use the knowledge I gained there on a wee…

Day 20: bell hooks.
Despite having decided to continue to 30, number 20 feels important, and hooks gets the spot in part because I haven't yet included a non-fiction feminist author, which feels like an obvious thing to include on such a list. The one category of author being bumped out of the first 20 here is anime writers, but I'll follow up with one of them, along with more academics and mangaka who I've been itching to include.
In any case, hooks is absolutely legendary as a feminist writer for good reason, and as a teacher I've especially appreciated her writing on pedagogy like "Teaching to Transgress: Education as the Practice of Freedom" and "Teaching Critical Thinking: Practical Wisdom". These have challenged me to teach at a higher level, and while I'm not sure I've completely succeeded, they're important to me. They also pair well with Paolo Friere's "Pedagogy of the Oppressed", but hooks always seems to be focused on very practical advice and it's incredibly direct in her writing, even though her advice isn't always straightforward to implement. In fact, that's one of the things I value about her writing: when the truth is complicated or the real work is messy interpersonal relationships that need to be negotiated with each student, she's not afraid to say so and give good advice for navigating those waters instead of trying to dispense simple-seeming platitudes or formulas for success that paper over the deeper issues. Her concern has always been truth, rather than simplicity or audience comfort and the popularity it might seem to entail, which I think is part of why her legacy endures so well.
#20AuthorsNoMen
#30AuthorsNoMen

Incremental Summarization for Customer Support via Progressive Note-Taking and Agent Feedback
Yisha Wu (Mia), Cen (Mia), Zhao, Yuanpei Cao, Xiaoqing Su, Yashar Mehdad, Mindy Ji, Claire Na Cheng
https://arxiv.org/abs/2510.06677

Incremental Summarization for Customer Support via Progressive Note-Taking and Agent Feedback
We introduce an incremental summarization system for customer support agents that intelligently determines when to generate concise bullet notes during conversations, reducing agents' context-switching effort and redundant review. Our approach combines a fine-tuned Mixtral-8x7B model for continuous note generation with a DeBERTa-based classifier to filter trivial content. Agent edits refine the online notes generation and regularly inform offline model retraining, closing the agent edits feedba…

Curious that whenever someone shows me “the cool #AI flow” they built that’s supposed to be impressive, the conversation goes the same way:
Stage 1: “But you don’t understand. You don’t like AI because you haven’t used it right. Let me show you how much you can do it with.”
Stage 2: “Here are the steps in the flow and the instructions I feed to this agent / custom GPT / Claude project. I tell it to do X, reference document Y, and aim for Z.”
Stage 3: “Now, let me show you the results it gives.”
*Writes task, presses to run the prompt.*
Stage 4: “Umm sorry it’s taking a while. It’s fast but not instant. And by the way, the prompt isn’t perfect, you can definitely make it better. I just threw this together real quick the other day. It makes some mistakes, but it’s really good.”
Stage 5: “Uuuuuuh actually don’t look at the output.” *scrolls or stops screen share or pulls device away.*
“You know it’s already doing so well, if I do more prompt engineering it will get really good but I need to give it better instructions. And it ran just fine last night, I don’t know what’s up with it. And this is a cheap model, if we use another model it will be better.”
Stage 6: “You know, you really shouldn’t judge this so much. The technology will improve, it will get there sooner than you know and then you’ll regret not trying it sooner.”
So curious that this keeps happening 🤷‍♀️
#LLMs #work #tech #AIBubble

Wow. I've dealt with various toxic personalities in software development, but a good portion of the time those toxic personalities were at least extremely knowledgeable in their (often, very limited) domain.
AI, however, seems to be enabling toxic personalities *who are completely clueless*. Impressive!
https://github…