Both GPT-4.1 and o4-mini are significantly better at image comprehension than the same models in the previous generations.

I know we like to clown on LinkedIn over here, but there's a lot of revenue to be generated if you know how to play the game...9 things I wish I knew when I started posting (full guide):

We recently ran a PoC to test inference workloads using DeepSeek R1 on accelerated hardware. We found: - Output speed was 10x faster than OpenAI (!!) - Output quality was perceptibly worse compared to GPT-4.1, but not drastically so - Cost would have been comparable to OpenAI's

Which is easier to read, with or without colors? 1️⃣ With colors 2️⃣ Without colors

The Eraser Slack community has grown to 3,000 builders – and counting. Thank you to everyone who continues to share, collaborate, and build with us. We’re grateful for your support.

May was quietly our biggest month yet – no big launch, no one logo carrying the month, just steady momentum as the AI revolution keeps rolling. The only thing slowing us down? Finding more great product engineers. Know someone who’d be a fit at Eraser? Send them our way!

Here's a clear visual explanation of why GCP was down for 2.5 hours last Thursday: 🚨 First, our service (Eraser) was affected by this outage as well. We're sorry about our outage. 🔍 What caused the outage? – GCP relies on a system called Service Control to validate API

A wholesome online interaction today that put a smile on my face: I hopped on a quick 15-minute reference call with another startup founder to chat about a candidate we're interviewing. After briefly sharing about our startups, we mostly discussed the candidate. A few minutes

👀 Coming soon to Eraser Who knows what these are called?

What do you call these customers? The kind of customers that: – Totally understands your product vision, and viscerally feels the pain you're solving for – Can accurately map out your competitive landscape and has tried every product under the sun – Has tinkered to create a

🤔 Which is easier to follow? Left-aligned vs. center-aligned?

We saw record growth in May – and then doubled it in June! Things are really taking off at Eraser. If you're a talented product engineer excited about building a fast-growing AI application – or know someone who fits the bill – please reach out at [email protected]. And if

There's a feature I would have been thrilled for a junior engineer to build in 3 days. Yesterday, using Claude Code, I finished it myself in just 6 hours. I experienced a similar revelation 6 months ago with Cursor, where suddenly I could fix bugs and spin up internal tools

In my recent conversations with enterprise IT customers, @Glean is emerging as a popular choice for AI agent orchestration. – Glean = Enterprise Q&A app → Headless AI orchestration layer (RAG + tool calling) – UI layer = Lovable app, MS Teams bots, GSuite integrations –

About to go on a tech talk in front of 200+ architects to discuss AI diagramming, wish me luck! Reach out if you want to book me for a talk for your own Architect or GenAI community!

I'm still amazed at how much Claude Code feels like a no-brainer at $100–$200/month. A few quick observations: 1️⃣ New price anchor: This pricing sets a completely new standard, breaking away from the traditional SaaS range of $10–$30/month that’s been the norm for the past two

Immediate thoughts after trying the new OSS 120B model: – Intelligence feels on par with o4/o3-mini – Already live on OpenRouter at just ~15-20% the price of o4/o3-mini – Should unlock a LOT of exciting new use cases for developers – OpenAI open sourcing this model strongly