<rss version="2.0" xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>Hacker News: w01fe</title><link>https://news.ycombinator.com/user?id=w01fe</link><description>Hacker News RSS</description><docs>https://hnrss.org/</docs><generator>hnrss v2.1.1</generator><lastBuildDate>Fri, 08 May 2026 14:28:54 +0000</lastBuildDate><atom:link href="https://hnrss.org/user?id=w01fe" rel="self" type="application/rss+xml"></atom:link><item><title><![CDATA[New comment by w01fe in "Natural Language Autoencoders: Turning Claude's Thoughts into Text"]]></title><description><![CDATA[
<p>This is incorrect.  In the process of producing each token, activations are produced at each layer which are made available to future token production processes via the attention mechanism.  The overall depth of computations that use this latent information without passing through output tokens is limited to the depth of the network, but there has been ample evidence that models can do limited "planning" and related capabilities purely in this latent space.</p>
]]></description><pubDate>Thu, 07 May 2026 21:21:29 +0000</pubDate><link>https://news.ycombinator.com/item?id=48055227</link><dc:creator>w01fe</dc:creator><comments>https://news.ycombinator.com/item?id=48055227</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=48055227</guid></item><item><title><![CDATA[New comment by w01fe in "Embodiment is indispensable for AGI"]]></title><description><![CDATA[
<p>I think this ignores the fact that an agent can be meaningfully embodied in the internet, using APIs for sensors and actuators.  OpenAI's training of large language models with reinforcement learning, recent retrieval augmented models and "chain of thought" reasoning are all meaningful steps in this direction in my opinion.</p>
]]></description><pubDate>Tue, 07 Jun 2022 22:12:42 +0000</pubDate><link>https://news.ycombinator.com/item?id=31660678</link><dc:creator>w01fe</dc:creator><comments>https://news.ycombinator.com/item?id=31660678</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=31660678</guid></item><item><title><![CDATA[Designing a Framework for Conversational Interfaces with PL Plus Constraints]]></title><description><![CDATA[
<p>Article URL: <a href="https://www.microsoft.com/en-us/research/group/msai/articles/designing-a-framework-for-conversational-interfaces/">https://www.microsoft.com/en-us/research/group/msai/articles/designing-a-framework-for-conversational-interfaces/</a></p>
<p>Comments URL: <a href="https://news.ycombinator.com/item?id=29788193">https://news.ycombinator.com/item?id=29788193</a></p>
<p>Points: 1</p>
<p># Comments: 0</p>
]]></description><pubDate>Mon, 03 Jan 2022 22:51:51 +0000</pubDate><link>https://www.microsoft.com/en-us/research/group/msai/articles/designing-a-framework-for-conversational-interfaces/</link><dc:creator>w01fe</dc:creator><comments>https://news.ycombinator.com/item?id=29788193</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=29788193</guid></item><item><title><![CDATA[New comment by w01fe in "Ask HN: Who is hiring? (January 2022)"]]></title><description><![CDATA[
<p>Microsoft (Semantic Machines) | Senior Software Engineer | Berkeley, CA or Boston, MA or Redmond, WA or fully REMOTE | Full Time<p>The Semantic Machines group is bringing next-generation natural language processing (NLP) technologies to products used by hundreds of millions of people worldwide. You can learn more about how Microsoft is using this technology to create entirely new kinds of user experiences here: <a href="http://semanticmachines.com/" rel="nofollow">http://semanticmachines.com/</a>.<p>At the core of our platform is a new programming language, designed to support programs structured like human commands which are predicted by machine learning models. This language includes ideas from functional and logic/constraint programming, as well as novel features for meta-computation and introspection. We’re looking for engineers to work alongside product and research teams to help guide the evolution of this platform, including improvements to the core programming language, runtime, constraint system, and tooling.<p>The ideal candidate should be passionate about designing and evolving programming languages and/or practical formal reasoning systems, supporting users with high quality tools, and working on a rapidly evolving product-driven platform. No experience with machine learning or natural language processing is required – we’d love to work with people who are excited about the promise of these technologies and the opportunity to make them more accessible, regardless of their previous exposure to them.<p>Learn more: <a href="https://www.microsoft.com/en-us/research/group/msai/articles/designing-a-framework-for-conversational-interfaces/" rel="nofollow">https://www.microsoft.com/en-us/research/group/msai/articles...</a><p>Apply: <a href="https://careers.microsoft.com/us/en/job/1215447/Senior-Software-Engineer" rel="nofollow">https://careers.microsoft.com/us/en/job/1215447/Senior-Softw...</a></p>
]]></description><pubDate>Mon, 03 Jan 2022 22:49:19 +0000</pubDate><link>https://news.ycombinator.com/item?id=29788160</link><dc:creator>w01fe</dc:creator><comments>https://news.ycombinator.com/item?id=29788160</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=29788160</guid></item><item><title><![CDATA[New comment by w01fe in "Ask HN: Best Engineering Blog Posts?"]]></title><description><![CDATA[
<p>We put a lot of work into this post to explain some of the deeper problems we are trying to address in our conversational AI framework:<p><a href="https://www.microsoft.com/en-us/research/group/msai/articles/designing-a-framework-for-conversational-interfaces/" rel="nofollow">https://www.microsoft.com/en-us/research/group/msai/articles...</a></p>
]]></description><pubDate>Sat, 01 Jan 2022 19:37:03 +0000</pubDate><link>https://news.ycombinator.com/item?id=29761833</link><dc:creator>w01fe</dc:creator><comments>https://news.ycombinator.com/item?id=29761833</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=29761833</guid></item><item><title><![CDATA[New comment by w01fe in "Ask HN: What companies are using probabilistic programming?"]]></title><description><![CDATA[
<p>At Semantic Machines [0] we rely heavily on probabilistic programming to build state-of-the-art dialogue systems.  In particular, we use a library called PNP (probabilistic neural programming) on top of Dynet to allow us to express structured prediction problems in a simple and elegant form.  If there are questions I am happy to elaborate to the extent I can.  (Also, we are hiring!  My email is jwolfe@.)<p>[0] <a href="http://www.semanticmachines.com/" rel="nofollow">http://www.semanticmachines.com/</a></p>
]]></description><pubDate>Mon, 04 Jun 2018 05:16:49 +0000</pubDate><link>https://news.ycombinator.com/item?id=17224868</link><dc:creator>w01fe</dc:creator><comments>https://news.ycombinator.com/item?id=17224868</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=17224868</guid></item><item><title><![CDATA[Microsoft spells out its plans for Semantic Machines]]></title><description><![CDATA[
<p>Article URL: <a href="https://venturebeat.com/2018/05/21/microsoft-spells-out-its-plans-for-semantic-machines/amp/">https://venturebeat.com/2018/05/21/microsoft-spells-out-its-plans-for-semantic-machines/amp/</a></p>
<p>Comments URL: <a href="https://news.ycombinator.com/item?id=17123634">https://news.ycombinator.com/item?id=17123634</a></p>
<p>Points: 1</p>
<p># Comments: 0</p>
]]></description><pubDate>Tue, 22 May 2018 03:59:55 +0000</pubDate><link>https://venturebeat.com/2018/05/21/microsoft-spells-out-its-plans-for-semantic-machines/amp/</link><dc:creator>w01fe</dc:creator><comments>https://news.ycombinator.com/item?id=17123634</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=17123634</guid></item><item><title><![CDATA[New comment by w01fe in "Schema graduates from alpha with support for Clojure test data generation"]]></title><description><![CDATA[
<p>Configurability was a major design goal.  Check out the test for examples of how to insert your own leaf generators or wrappers:<p><a href="https://github.com/Prismatic/schema/blob/master/test/clj/schema/experimental/generators_test.clj" rel="nofollow">https://github.com/Prismatic/schema/blob/master/test/clj/sch...</a><p>Does that seem like it will meet your needs?</p>
]]></description><pubDate>Tue, 01 Sep 2015 23:32:25 +0000</pubDate><link>https://news.ycombinator.com/item?id=10156524</link><dc:creator>w01fe</dc:creator><comments>https://news.ycombinator.com/item?id=10156524</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=10156524</guid></item><item><title><![CDATA[New comment by w01fe in "Schema graduates from alpha with support for Clojure test data generation"]]></title><description><![CDATA[
<p>Author here, would love to hear feedback, and more than happy to answer questions</p>
]]></description><pubDate>Tue, 01 Sep 2015 18:07:13 +0000</pubDate><link>https://news.ycombinator.com/item?id=10154477</link><dc:creator>w01fe</dc:creator><comments>https://news.ycombinator.com/item?id=10154477</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=10154477</guid></item><item><title><![CDATA[Schema graduates from alpha with support for Clojure test data generation]]></title><description><![CDATA[
<p>Article URL: <a href="http://blog.getprismatic.com/schema-1-0-released/">http://blog.getprismatic.com/schema-1-0-released/</a></p>
<p>Comments URL: <a href="https://news.ycombinator.com/item?id=10154400">https://news.ycombinator.com/item?id=10154400</a></p>
<p>Points: 15</p>
<p># Comments: 3</p>
]]></description><pubDate>Tue, 01 Sep 2015 17:57:47 +0000</pubDate><link>http://blog.getprismatic.com/schema-1-0-released/</link><dc:creator>w01fe</dc:creator><comments>https://news.ycombinator.com/item?id=10154400</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=10154400</guid></item><item><title><![CDATA[Prismatic Pivots to B2B, Packaging Its Content Interest Graph into APIs]]></title><description><![CDATA[
<p>Article URL: <a href="http://techcrunch.com/2015/07/20/prismatic-pivots/">http://techcrunch.com/2015/07/20/prismatic-pivots/</a></p>
<p>Comments URL: <a href="https://news.ycombinator.com/item?id=9918917">https://news.ycombinator.com/item?id=9918917</a></p>
<p>Points: 2</p>
<p># Comments: 1</p>
]]></description><pubDate>Mon, 20 Jul 2015 20:42:35 +0000</pubDate><link>http://techcrunch.com/2015/07/20/prismatic-pivots/</link><dc:creator>w01fe</dc:creator><comments>https://news.ycombinator.com/item?id=9918917</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=9918917</guid></item><item><title><![CDATA[Interest Graph Topic Feeds API]]></title><description><![CDATA[
<p>Article URL: <a href="http://blog.getprismatic.com/recent-urls/">http://blog.getprismatic.com/recent-urls/</a></p>
<p>Comments URL: <a href="https://news.ycombinator.com/item?id=9298778">https://news.ycombinator.com/item?id=9298778</a></p>
<p>Points: 1</p>
<p># Comments: 0</p>
]]></description><pubDate>Tue, 31 Mar 2015 18:29:40 +0000</pubDate><link>http://blog.getprismatic.com/recent-urls/</link><dc:creator>w01fe</dc:creator><comments>https://news.ycombinator.com/item?id=9298778</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=9298778</guid></item><item><title><![CDATA[New comment by w01fe in "Migrations and Future Proofing"]]></title><description><![CDATA[
<p>Author here, would love feedback on this, and also happy to answer any questions.</p>
]]></description><pubDate>Wed, 24 Sep 2014 21:07:31 +0000</pubDate><link>https://news.ycombinator.com/item?id=8364160</link><dc:creator>w01fe</dc:creator><comments>https://news.ycombinator.com/item?id=8364160</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=8364160</guid></item><item><title><![CDATA[Migrations and Future Proofing]]></title><description><![CDATA[
<p>Article URL: <a href="https://github.com/Prismatic/eng-practices/blob/master/swe/Migrations_and_future_proofing.md">https://github.com/Prismatic/eng-practices/blob/master/swe/Migrations_and_future_proofing.md</a></p>
<p>Comments URL: <a href="https://news.ycombinator.com/item?id=8363027">https://news.ycombinator.com/item?id=8363027</a></p>
<p>Points: 74</p>
<p># Comments: 12</p>
]]></description><pubDate>Wed, 24 Sep 2014 18:03:06 +0000</pubDate><link>https://github.com/Prismatic/eng-practices/blob/master/swe/Migrations_and_future_proofing.md</link><dc:creator>w01fe</dc:creator><comments>https://news.ycombinator.com/item?id=8363027</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=8363027</guid></item><item><title><![CDATA[New comment by w01fe in "Om Sweet Om: functional front-end engineering with ClojureScript and React"]]></title><description><![CDATA[
<p>Yep, that sounds about right.  For us, API communication and some global events flow through channels into the app state, where it's distributed to components.  For routing we've built up a library around secretary, where we write down a map of uris, handlers, and metadata (can be a modal, etc).  Info about the current page and such goes into the app state and that drives the transitions between page components.</p>
]]></description><pubDate>Tue, 17 Jun 2014 22:16:59 +0000</pubDate><link>https://news.ycombinator.com/item?id=7907373</link><dc:creator>w01fe</dc:creator><comments>https://news.ycombinator.com/item?id=7907373</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=7907373</guid></item><item><title><![CDATA[New comment by w01fe in "Om Sweet Om: functional front-end engineering with ClojureScript and React"]]></title><description><![CDATA[
<p>We're still figuring out exactly where the boundary lies, but our current approach is much like yours -- view state that's purely local to a component goes in local state, and everything else goes in app state.  This feels more natural, but means you lose some of the nice benefits of things like replay/undo for this local state.<p>For client/server, we've started building up some nice abstractions around our API (like fetching more data in a generic paginated list stored in a cursor), but the write side is still mostly manual.  We're trying to use the same API for iOS and web, which probably prevents us from doing some fancier things that continue the same architecture back into the server.</p>
]]></description><pubDate>Tue, 17 Jun 2014 22:12:20 +0000</pubDate><link>https://news.ycombinator.com/item?id=7907349</link><dc:creator>w01fe</dc:creator><comments>https://news.ycombinator.com/item?id=7907349</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=7907349</guid></item><item><title><![CDATA[New comment by w01fe in "Om Sweet Om: functional front-end engineering with ClojureScript and React"]]></title><description><![CDATA[
<p>No, you are still free to organize your application state however you wish.  For example, our app dynamically pulls in content as you browse from feed to feed and use infinite scroll.  It's up to you to decide what to fetch when, and what to cache/discard as the user navigates around.</p>
]]></description><pubDate>Tue, 17 Jun 2014 22:06:44 +0000</pubDate><link>https://news.ycombinator.com/item?id=7907328</link><dc:creator>w01fe</dc:creator><comments>https://news.ycombinator.com/item?id=7907328</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=7907328</guid></item><item><title><![CDATA[New comment by w01fe in "Om Sweet Om: functional front-end engineering with ClojureScript and React"]]></title><description><![CDATA[
<p>Engineer @ Prismatic here.  I'm happy to answer any questions about the post or our experiences with the rewrite.</p>
]]></description><pubDate>Tue, 17 Jun 2014 20:23:48 +0000</pubDate><link>https://news.ycombinator.com/item?id=7906777</link><dc:creator>w01fe</dc:creator><comments>https://news.ycombinator.com/item?id=7906777</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=7906777</guid></item><item><title><![CDATA[New comment by w01fe in "Value Types for Java – A sketch of proposed enhancements"]]></title><description><![CDATA[
<p>Also perhaps interesting for comparison, a Clojure library called Vertigo that emulates C structs with high performance on top of byte buffers:<p><a href="https://github.com/ztellman/vertigo" rel="nofollow">https://github.com/ztellman/vertigo</a></p>
]]></description><pubDate>Fri, 02 May 2014 16:03:58 +0000</pubDate><link>https://news.ycombinator.com/item?id=7686472</link><dc:creator>w01fe</dc:creator><comments>https://news.ycombinator.com/item?id=7686472</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=7686472</guid></item><item><title><![CDATA[New comment by w01fe in "Claypoole: Threadpool Tools for Clojure"]]></title><description><![CDATA[
<p>I'm really excited about this library -- thanks Leon and Climate for releasing it!</p>
]]></description><pubDate>Tue, 25 Feb 2014 18:26:11 +0000</pubDate><link>https://news.ycombinator.com/item?id=7299404</link><dc:creator>w01fe</dc:creator><comments>https://news.ycombinator.com/item?id=7299404</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=7299404</guid></item></channel></rss>