Ian

3.1K posts

Ian banner
Ian

Ian

@ianbroom

Revolutionising mobile in enterprise with @FlipletApp, working with global clients and geeking out about startups, technology and mobile.

London, UK Katılım Ağustos 2008
364 Takip Edilen682 Takipçiler
Ian
Ian@ianbroom·
@zackbshapiro Recently I migrated from Claude to OpenAI as I found the quality of work higher and it more diligent doing the whole task. Have you compared it to Codex desktop app? I wonder what the gap is between the products especially as the models are constantly improving.
English
0
0
0
20
Ian
Ian@ianbroom·
Management offsites can be stressful. Making sure everybody knows where they need to be, what they need to be doing, and what's expected of them is critical information. Heather built the solution without IT or design assistance too!
Fliplet: Rapidly create apps@flipletapp

Red Dog Media built a professional event app for their leadership retreat — without IT support or app store approvals. ✔Centralized agenda & logistics ✔ Real-time updates ✔ Built by a non-technical team member ✔ Shared with attendees before the event events.fliplet.com/case-studies/r…

English
0
0
0
22
Peter Yang
Peter Yang@petergyang·
My first impressions with Codex (yeah I only started seriously using it this weekend): 1. It can find and fix bugs that Claude Code introduced (!) 2. It is not very good at UX/frontend - maybe there's a skill that can help? 3. $20/month seems like great value
English
116
13
639
71.7K
Ian
Ian@ianbroom·
Why do people keep asking questions that ai knows the answer to? Must be hoping other users will use ai to send them the answer.
English
0
0
0
8
Ian
Ian@ianbroom·
It bothers me that @claudeai code fails to 1. find its settings file 2. creates skills in the wrong location 3. configures MCP in the wrong location. It's frustrating having to restart claude code and find settings it just configured don't work. Am I missing something?
English
1
1
2
124
Ian
Ian@ianbroom·
@BosonJoe How do you chat with it from anywhere?
English
1
0
0
686
Joe Muller
Joe Muller@BosonJoe·
Clawdbot is basically claude code in a folder with a few markdown files The docs make the whole thing more confusing than it has to be Here's Gus, my personal AI assistant running on my macbook. It has slack and google workspace skills and a CLAUDE.md file with basic operating instructions And I can chat with it from anywhere
Joe Muller tweet media
English
4
2
51
24.4K
Ian
Ian@ianbroom·
@levelsio I appreciate you sharing life hacks like this.
English
0
0
0
4
@levelsio
@levelsio@levelsio·
Guys we have 2 glass airfryers now Can make 2x 500g 🥩 ribeye at the same time (1kg steak = 250g protein) Or 1x 500g 🥩 ribeye + 1x 500g 🥦 vegetables Also rice cooker on right and SodaStream sparkling water maker on left (I press it 6x to get turbo water, 1x is too weak)
@levelsio tweet media@levelsio tweet media@levelsio tweet media
@levelsio@levelsio

It's so good I'm going to buy a second Ninja Crispi airfryer so I can cook frozen meat in one and frozen vegetables in the second one This + my ceramic rice cooker = Rice + meat + veg, all done without almost any work after 15 minutes (and no microplastics!) Not affiliated

English
218
31
1.6K
754.4K
Ian
Ian@ianbroom·
Google released generative UI. OpenAI released Apps. I think Gen UI will win for 2 reasons: 1. No one wants to build apps proactively against a proprietary standard 2. Gen UI is likely to be more dynamic and evolve as the model improves
Google Research@GoogleResearch

Generative UI is a capability where an AI model generates dynamic, visual layouts & interactive interfaces, such as web pages, games, & apps. Today with the launch of Gemini 3, you can experience our novel implementation of generative UI in @GeminiApp & Google Search. 🧵↓ & blog →goo.gle/4r6Gp1V

English
0
1
1
41
Ian
Ian@ianbroom·
@zarazhangrui @karpathy Great idea. I do this a lot with gemini app but I have to write the prompt manually each time.
English
1
0
0
355
Zara Zhang
Zara Zhang@zarazhangrui·
@karpathy I do the same with long YouTube videos (such as your interview with Dwarkesh), and developed a tool for this use case: tldw.us
English
9
2
52
11.9K
Andrej Karpathy
Andrej Karpathy@karpathy·
I’m starting to get into a habit of reading everything (blogs, articles, book chapters,…) with LLMs. Usually pass 1 is manual, then pass 2 “explain/summarize”, pass 3 Q&A. I usually end up with a better/deeper understanding than if I moved on. Growing to among top use cases. On the flip side, if you’re a writer trying to explain/communicate something, we may increasingly see less of a mindset of “I’m writing this for another human” and more “I’m writing this for an LLM”. Because once an LLM “gets it”, it can then target, personalize and serve the idea to its user.
English
596
1.1K
13.3K
2.9M
Ian
Ian@ianbroom·
@flipletapp @tonytlwu I can't wait to speak with Emma and Grace from Mills & Reeve to learn what they've been using Fliplet for
English
0
0
0
13
Ian
Ian@ianbroom·
@flipletapp I am looking forward to presenting next week and sharing our plans for the future.
English
0
0
0
8
Ian
Ian@ianbroom·
Dear @OpenAI I have been stuck in a loop with your support team for weeks. I cannot upgrade. I was testing the Team plan but the inability to share conversations publicly was a deal breaker so I downgraded to free and I now cannot upgrade again. Help!
Ian tweet media
English
0
0
0
21
Ian
Ian@ianbroom·
@HanchungLee Gemini can produce detailed summaries
English
0
0
0
32
Han
Han@HanchungLee·
i dont listen to podcasts; found them to be a complete waste of time. until i found the chinese tech podcasts.
English
60
185
3.9K
476.6K
Ian
Ian@ianbroom·
@peternixey @bryan_johnson This is really interesting especially for content that doesn’t require an exact replication although you’re very dependent on how the model interprets the description. I wonder what happens if you ask for far more detail with the goal of a direct reproduction?
English
2
0
0
29
Peter Nixey
Peter Nixey@peternixey·
They say a picture speaks 1,000 words. But it turns out that you can recreate most of a picture with just 500. Which kind of blows my mind. Here's 5 examples of compressing an image to under 600 words and then recreating it again using AI. Let's start with @bryan_johnson 👇🏻
English
2
0
1
683
Ian retweetledi
Fliplet: Rapidly create apps
Fliplet: Rapidly create apps@flipletapp·
Congratulations to the team at @foxwilliams for winning our App of the Month award! Their Partners app gives the firm’s partners an events app to take notes, see their agenda and important information when attending conferences and other events.
Fliplet: Rapidly create apps tweet media
English
0
1
1
44
Ian
Ian@ianbroom·
@lovable It would be great to understand what other libraries loveable is good at using
English
0
0
0
93
Lovable
Lovable@Lovable·
With Lovable, you can easily create 3D graphs, timelines, and interactive node maps by simply specifying in the prompt that it should use Vis.js
English
2
14
180
10.6K
Ian
Ian@ianbroom·
@WisprAI I have installed your app but it just keeps loading and closing without any errors. Any idea what's going on? I couldn't find a FAQ or support site
English
1
0
0
18