Warsaw Meeting Tuesday 4th

How does product adjust?

how does arch adjust?

Eng needs to have more of a product mindset and product needs to be building more

Full AI developed, but starting from scratch with less legacy dependencies

Better auth than we currently have, stripe integration for self service montetization with plug ins

Use AI iterations to reduce product rework by front loading discovery and find the best solution

Wants it to be in Vue and Postgres so hand over is easiest for ENG

Eng skeptical that the hand over will be efficient 

25% of time on code, is the optimization the actual issue

PRoduct going through vercel iterations allows for easier dog fooding to make requirements more clear and accurate because there are miles on ai

3/4 of work is before the coding

Ben "fix product first" 

Both sides need to solve for their inefficiencies (pre dev and dev)

interally test to see if it creates value with layer it on 

Split teams into smaller domains and smaller head count to reduce communication overhead. Have them focus on bugs and TD 

Bigger companies like atlassian are spending 25-40% on TD

Small teams like 3-4 with 1 person dedicated to TD, bugs and public channles

2-3 others just focused on new features

Nick has bots that monitor logs that go and fix errors. Why aren't we doing it in Sentry

Certain features are legacy and holding us back

With the knowledge known now how quickly could W2 reimplement tasks in? 

Do we have configured skills to make sure ai agent are consistent and effective across teams

Using ai to go from functional requirements then iterate on technical decisions to then develop technical specs to make development of production code with ai in one massive prompt

Going through the feedback loop internally with prototypes rather than with production release

Nick wants hand off through skills to make the code 80% there with engineering polish makes translating to prod easy

Meeting minutes prototype better than bidding because Eli is closer to the source of feedback

PM's become more engineers and engineers get more producty

Everyone needs to be in the mindset of iterations and automation to get better features

Eli vibe codes and merges 20k lines of code

Need a plan to bridge the gap from prototype to production code

why does engineering need to reverse engineer rather than integrate the code

Pre-development from Nick's perspective is solved for. Prototype and kick the tires until were happy

Unified skills to make the code and tech stack as aligned and effective as possible

Short term that does not matter but long term that gap will close so should be working towards the prototype can eventually be used with minimal hand off 

Get the design system into the prototypes and other changes to make it more effective and efficient

The team discussed the rapid development and deployment of standalone AI-driven applications within the industry, noting the implementation of an isolated app prototype for change orders. No formal decisions or action items were established.