Loading Sangam…
11 discussions on Sangam.
Looking for an AI tool for my business...
Struggling to get first customers...
Need feedback on my startup...
Not clickbait — measured across six weeks of tickets. The workflow: spec in plain English first, let the agent draft, re...
Open models have closed the quality gap for a lot of tasks, but every time I run the numbers, hosted closed models still...
Did the math and I'm at $87/month: ChatGPT Plus, Claude Pro, Cursor, and Midjourney. Each one earns its keep but the tot...
Building a doc-QA feature for a client with almost no infra budget. Current setup: chunk at ~500 tokens, small embedding...
Everyone says wrappers are dead, yet I keep meeting people running $10K MRR 'ChatGPT for X' apps. My take: the wrapper w...
Benchmarks are one thing, day-to-day agent work is another. In my experience Claude follows multi-step instructions more...
Every raw draft has the same tells: 'delve', 'moreover', rule-of-three sentences, and a conclusion that starts with 'In ...
I've bounced between both for three months. Cursor's agent mode feels more reliable on large refactors, but Windsurf und...