Paul Earnden | The HubSpot Agency Leader (@paulearnden) header
Paul Earnden | The HubSpot Agency Leader (@paulearnden) profile picture

MD of @fuelius_agency, a global Elite @HubSpot Partner.

England, UK
Joined August 2014
Photo by Paul Earnden | The HubSpot Agency Leader (@paulearnden)Photo by Paul Earnden | The HubSpot Agency Leader (@paulearnden)Photo by Paul Earnden | The HubSpot Agency Leader (@paulearnden)Photo by Paul Earnden | The HubSpot Agency Leader (@paulearnden)Photo by Paul Earnden | The HubSpot Agency Leader (@paulearnden)Photo by Paul Earnden | The HubSpot Agency Leader (@paulearnden)Photo by Paul Earnden | The HubSpot Agency Leader (@paulearnden)Photo by Paul Earnden | The HubSpot Agency Leader (@paulearnden)Photo by Paul Earnden | The HubSpot Agency Leader (@paulearnden)Photo by Paul Earnden | The HubSpot Agency Leader (@paulearnden)
Imagine selling change management using a slide deck loooool ngmi
1
1
182
Yep
if you are not skillsmaxxing with o3 at minimum 3 hours every day, ngmi
2
329
About 10 months ago I started trying to vibe code a product in ruby, but LLMs weren't really great at code. Since then i've done a tremendous amount of research, and trial and error and I am absolutely stunned by Gemini 2.5 Pro. I am probably 25% through a hobbyist side project @HubSpot app development using python and specifically @scikit_learn for machine learning - and Gemini is literally one-shotting everything, flawlessly. Still using @pycharm by @jetbrains who were kind enough to gift me a license. This is flying now, hope to launch it in April as a passion project! Tech stack = @ThePSF @djangoproject @scikit_learn #vibecode #buildinpublic
1
225
Paul Earnden | The HubSpot Agency Leader retweeted
The future belongs to idea guys who can just do things
6
26
1
1,060
358,367
Paul Earnden | The HubSpot Agency Leader retweeted
No longer hiring junior or even mid-level software engineers. Our tokens per codebase: Gumroad: 2M Flexile: 800K Helper: 500K Iffy: 200K Shortest: 100K Both Claude 3.5 Sonnet and o3-mini have context windows of 200K tokens, meaning they can now write 100% of our Iffy and Shortest code if prompted well. It won’t be long until AI will be writing all the code for Helper, Flexile, and Gumroad. My guess is by the end of 2026. Our new process: 1. Sit and chat about what we need to build, doing research with Deep Research as we go. 2. Have AI record everything and turn it into a spec. 3. Clean up the spec, adding any design requirements / other nuances. 4. Have Devin code it up. 5. QA, merge, (auto-)deploy to prod.
528
914
447
12,721
3,702,993
Hey @AvantiWestCoast when are you going to bring @Starlink to your trains? Great train infrastructure in UK let down by 1990s internet. Cc @elonmusk
193
Tim Cook calling the new Apple Watch a “beautiful new design” is stretching a bit. #AppleEvent
1
387