#ai-ml
it's interesting that recommender systems have become colloquially known as "algorithms" -- when I think that term implies a certain amount of strict mathematical correctness ML/AI systems really don't have lol
January 18, 2026 at 9:08 PM
The way this works: you find new TV

1. Select shows you like, hide ones you dont
2. It finds writers from them
3. It finds their other work
4. It weighs it by season count, star rating, number of writer overlap, and some other stuff
5. It recommends shows in a discover tab

no ML no AI just a graph
Ok I. have TestFlight links for my Role Call find TV thingy

Its just like the web version (in follow up) but saves things better cuz it doesn't just use LocalStorage from the web browser that can get wiped easily

testflight.apple.com/join/5GKarSrZ
TestFlight - Apple
Using TestFlight is a great way to help developers test beta versions of their apps.
testflight.apple.com
January 18, 2026 at 8:17 PM
mentally preparing myself to be kicked from an "artists who hate generative ai" group because i can't help but note that yeah, maybe traditional ML algorithms and LLMs combined with a human editor might be better at medical coding and charting than a human alone
January 18, 2026 at 8:10 PM
Solutions Architect, AI and ML - US, WA, Redmond Job educativ.net/jobs/job/51987...
January 18, 2026 at 7:24 PM
So I was just asked to explain why I hate AI so much, bc I work as an AI/ML systems architect.

Here‘s the TL;DR of it all: ai is neutral, it all depends on how you use it. People are 100% of the problem.
January 18, 2026 at 7:06 PM
bruh How to Read a ML Research Paper in 202 no thoughts just vibes

When I first kicked off reading ML research papers, I honestly thought something was wrong with me....

#AI #ArtificialIntelligence

🔗 https://trustmebro.pro/posts/how-to-read-a-ml-research-paper-in-2026
January 18, 2026 at 6:19 PM
Here’s a little example of my “role call” app to find tv by marking shows and it finds more shows that the writers from your corpus have done. It weighs it with a Bobby made algorithm for the for you tab. There’s no AI or ML just trees

If you want it DM and I’ll add you to beta
January 18, 2026 at 6:17 PM
AI/ML are driving a steep ramp in neural processing unit (NPU) design activity for everything from data centers to edge devices such as PCs and smartphones.
6 experts discuss how and why to optimize NPUs.
semiengineering.com/how-and-why-...

#NPU
How And Why To Optimize NPUs
PPA constraints need to be paired with real workloads, but they also need to be flexible to account for future changes.
semiengineering.com
January 18, 2026 at 5:41 PM
I'll be delivering #ArtificialIntelligence (#AI) for Project Managers: Leading an ML Team at Project World/BA World in Toronto on May 11. Hope to see you there. pmbaconferences.com/toronto/prog...
January 18, 2026 at 5:16 PM
In other news, Augustine's "On the Trinity" is the 50th book we've listed on our website. 🎉🥳

It's the fiftieth book we've used ML/AI techniques to newly translate from the Greek or newly revise/modernize from an older public domain edition.

appianwaypress.com (1/3)

#EarlyChristianity
Appian Way Press — Modern translations of ancient Christian works
Discover modern English translations of early Christian works using machine learning and translation technology. 50 books available.
appianwaypress.com
January 18, 2026 at 4:08 PM
»Moxie Marlinspike has a privacy-conscious alternative to ChatGPT« https:// techcrunch.com/2026/01/18/moxi e-marlinspike-has-a-privacy-conscious-alternative-to-chatgpt/?AIagent.at # AIagent # AI # ML # NLP # LLM # GenAI

Interest | Match | Feed
Origin
defcon.social
January 18, 2026 at 3:37 PM
The TechBeat: Why Data Quality Is Becoming a Core Developer Experience Metric (1/17/2026)

The Techbeat by HackerNoon provides fresh content from trending stories of the day, covering various topics such as AI, technology, and innovation. The newsletter features articles on go…
#hackernews #ml #news
The TechBeat: Why Data Quality Is Becoming a Core Developer Experience Metric (1/17/2026)
The Techbeat by HackerNoon provides fresh content from trending stories of the day, covering various topics such as AI, technology, and innovation. The newsletter features articles on governing and scaling AI agents, production-grade agent architecture, and patterns to avoid in AI agent deployment. It also includes articles on HR software for midsize companies, playbook for production ML, and the impact of AI on jobs. Additionally, the newsletter covers topics such as innovation and accountability in Web3 finance, RAG architectures, and ISO 27001 compliance tools. The articles are written by various authors, including @denisp, @stevebeyatte, and @chris127, and offer insights and expertise on the latest trends and technologies. The newsletter aims to provide valuable information and resources to its readers, and encourages them to share it with others who may be interested. The HackerNoon team also invites readers to contribute to the community by writing and sharing their own knowledge and experiences. The newsletter concludes by wishing its readers a great day and inviting them to explore more content on the HackerNoon platform. Overall, the newsletter offers a wide range of topics and insights, making it a valuable resource for anyone interested in technology and innovation. The articles are designed to be informative and engaging, and the newsletter is well-organized and easy to navigate.
hackernoon.com
January 18, 2026 at 3:19 PM
AI/ML Foundations for Absolute Beginners (AgenticAI + MLOps) Build a Solid Conceptual Foundation on Machine Learning, Large Language Models (LLMs) and Agentic AI along with MLOps ⏱️ Length: 4.3...

#StudyBullet-22 #coupons #development #Education #Free […]

[Original post on studybullet.com]
January 18, 2026 at 3:17 PM
»Under Musk, the Grok disaster was inevitable« https://www. theverge.com/column/863502/gro k-deepfake-musk?AIagent.at # AIagent # AI # ML # NLP # LLM # GenAI

Interest | Match | Feed
Origin
defcon.social
January 18, 2026 at 1:33 PM
»Under Musk, the Grok disaster was inevitable« https://www. theverge.com/column/863502/gro k-deepfake-musk?AIagent.at # AIagent # AI # ML # NLP # LLM # GenAI

Interest | Match | Feed
Origin
defcon.social
January 18, 2026 at 1:33 PM
ペットボトルの底はつまりこういうこと。このピッタリ具合が一番ドラマティックだった件🥹なお下にあるのはうちで一番高級なおレンズ様です🥹
January 18, 2026 at 11:12 AM
おれが前からブログに書いてること。

AIが発達すれば、AIのコードが正しいことが自動的に検証出来る必要が生まれる。その結果、静的型づけ以外無価値になる。
そしてその中でもLean4のように究極的な言語が使われるようになる。AIは、自身のコードが正しいことを証明するようになる。

Lean4は次世代のプログラミング言語です。

当面はRustがハネると思います。

github.blog/ai-and-ml/ll...
Why AI is pushing developers toward typed languages
AI is settling the “typed vs. untyped” debate by turning type systems into the safety net for code you didn’t write yourself.
github.blog
January 18, 2026 at 10:56 AM
あとオマケ。ミネラル水とかの綺麗な円筒型の500mLペットボトルの底だけ切り出してなんとなくフィルター径62mmのレンズの先端にはめてみたらめっちゃピッタリ🤤残念ながらなんにも写らないくらいぼやけるけど点光源に向けると面白いこと。イルミ撮れるうちに気づきたかった笑
#ブルスカ写真部
January 18, 2026 at 10:35 AM
(2026/1/18)(日)
こんばんは🤗✨はたぼ〜🌵️とaiトレイン🚃です🩵🚃 今日のお仕事💼を終えました✨️帰宅🏠️しています🙌️🤗️今日は晴れ🌤️で朝はひんやりでしたが昼間は寒さは緩み日差し🌤️の暖かさを感じる時もありました。終始防寒対策をした服装で調節しながら過ごし今日はこれまでに1773mlの水分補給🥛️をしました🍀️🤗️明日はお休み🌴でこの後は帰宅🏠️してゆるりと過ごそうと思います🍀️🤗️
【鐵道空想再現】
(名鉄6000系🚃️)②
January 18, 2026 at 8:30 AM
📢 Booz Allen Hamilton is #hiring a Scientific AI & ML Engineer!

💰 $77K - $176K
📍 Atlanta, GA

🔗 http://jbs.ink/qdgs1vMZFar6

#remotejob #remotework #wfh
January 18, 2026 at 7:06 AM
【ISTQB /JSTQB AI Tester 解説】機械学習(ML)ワークフローとは?

― ISTQB AI Tester試験で必須の基本プロセスを完全理解 ― ISTQB AI Tester Chapter 3では、機械学習(Machine Learning:ML)をどのような流れで構築・運用するのかという 「MLワークフロー(ML Workflow)」が重要なテーマとして扱われます。 機械学習モデルは、 作って終わりではなく、運用しながら改善し続けるものです。 その全体像を体系的に理解することが、AIテストにおいて不可欠になります。 MLワークフローの全体像…
【ISTQB /JSTQB AI Tester 解説】機械学習(ML)ワークフローとは?
― ISTQB AI Tester試験で必須の基本プロセスを完全理解 ― ISTQB AI Tester Chapter 3では、機械学習(Machine Learning:ML)をどのような流れで構築・運用するのかという 「MLワークフロー(ML Workflow)」が重要なテーマとして扱われます。 機械学習モデルは、 作って終わりではなく、運用しながら改善し続けるものです。 その全体像を体系的に理解することが、AIテストにおいて不可欠になります。 MLワークフローの全体像 MLワークフローは、大きく次の 2フェーズ に分かれます。 モデル開発(Model Development) デプロイ・利用・監視(Deploy, Use, Monitor) まずはモデルを作り、その後、実システムで使いながら継続的に改善していきます。 ① モデル開発フェーズ(Model Development) 1. 目的(Objective)の理解 最初に行うべき最重要ステップが 目的の明確化 です。 なぜこのMLモデルが必要なのか? 何を予測・分類・判断したいのか? ビジネス上のゴールは何か? これらをステークホルダーと合意し、 さらに 受け入れ基準(Acceptance Criteria) を定義します。 具体例 不正取引検知モデル →「不正検知率95%以上」「誤検知率5%以下」など 2. フレームワークの選定 目的・受け入れ基準・ビジネス要件に基づき、 AI/ML開発フレームワークを選択します。 TensorFlow PyTorch Scikit-learn など ISTQBでは、「なぜそのフレームワークを選んだのか」 という合理性が重要視されます。 3. アルゴリズムの選択・構築 次に、MLアルゴリズムを決定します。 既存ライブラリのアルゴリズムを使用 独自アルゴリズムを新規開発
testengineer.biz
January 18, 2026 at 6:43 AM
WTF did the Magic Fill feature in Numbers do to get pointed out in a bullet list of recent announcements …
It's a classic ML feature that Apple is known for more than an AI feature (not that that would be necessarily bad for learning formulas)
#theVerge
www.theverge.com/tech/863614/...
January 18, 2026 at 5:51 AM
📢 Gartner is #hiring a Lead Software Engineer- ML, NLP, Gen AI!

📍 Gurgaon, India

🔗 http://jbs.ink/84vRULIdxXr0

#remotejob #remotework #wfh
January 18, 2026 at 5:33 AM
📢 Gartner is #hiring a Manager, Software Engineering - ML, NLP, Gen AI!

📍 Gurgaon, India

🔗 http://jbs.ink/hljgKVcAntCd

#remotejob #remotework #wfh
January 18, 2026 at 5:28 AM