Curious to know the understanding about this - if the context window size of a LLM is ~1M - how does it understand and make changes to large codebases. Looking for answers beyond the traditional LLM onboarding ( claude.md, agent.md, divide into modules, TDD)
#codewithclaude
@UC_Assist I recently shifted to Jaipur and ordered a M2 for Rs 18,449. The order was delayed a lot and then the team said that they cannot deliver. I am near to pleading them to return the money atleast. My ticket got 'parked' atleast 10 times. Spoke to 5 executives. Help!!
Chat is a strong mental model for experiencing the value of AI. But it made me think: what other models exist? Here are 7 other categories I've identified, and I'd love for you to add more examples or insights in the comments.
#aiproductmanagement#productdesign
The right use of these AI experiences can accelerate the path to an 'aha' moment for users. They reduce friction, simplify onboarding, and make tasks feel less effortful
#Product#pmf#aiproduct
Leading indicator of a good problem to solve ? The effort to discover it is >>> than the effort to make it apparent
#productdesign#startups#ProblemSolving
In favour: Parkinson’s law - iterations will fill the time you will give them. Keeping it time bound will bring efficiency
Not in favour: Time bound effort might reduce discovery.
Thoughts?
#productmanagement
A PRD writing observation: Writing states and correspondingly covering behaviours based on states is an error prone area while writing PRD's
Some solutions: Visualise with diagram (flowchart), table
#productdevelopment#productdesign