i think what you want is a build step that acts like a “llm macro”, like you write comments or specs in the code, and the ai fills in the implementations automatically. its like literate programming but powered by llms.
theres a big gap now between interactive chatbots and fully automated code gen at build time. so a tool that integrates with the build pipeline and produces documented code would be a game changer.
also if it explains why it added code not in your specs that would improve trust a lot. this is possible with current models and some clever tooling but nobody cracked the ux yet.
personally ive tried a few hacks with rust macros and cargo steps to prototype this and it works ok but still rough around edges.
this will definitely be a hot area in next 1-2 years
inner work is way overlooked by founders and devs. we all grind on code and funding but ignore the mental stuff that burns us out.
a good guide would show how to spot when you need inner work, not just stress. teach simple ways to handle emotions, imposter syndrome, burnout, stuff that messes with focus and leadership.
also, quick practices for busy people, how inner work helps with team leading and deals. most think just “work harder” fixes it but really you gotta debug your brain first.
if you get this right, you dont just build a product, you build long term resilience.
i was in the same boat. loved coding but hated the 9-5 grind and had zero clue how to sell myself or my projects. what worked for me was ditching the idea of building something big from day one. instead i started fixing real small pains for people i knew. friends, family, coworkers.
i made dumb scripts that saved them minutes here and there, built small websites or automations that made their life easier. i never charged much, sometimes free, sometimes a small fee. but the key was getting feedback and iterating until they couldnt live without it.
then i slowly built a network, people started referring me, and jobs came from that. marketing wasnt some magic trick, it was just solving problems people already had and telling them how i do it.
also, find someone who loves marketing, partner with them. keep control over your code but let them handle the biz side.
freelance contract work can be stable and flexible too, dont feel pressured to build a product from scratch right away. grow your skills and network in parallel.
it wont be easy or fast but focus on real value, real people, and trust will follow.
I just commented elsewhere that your comment felt AI-written. Then I read this and I'm like "feels GenAI", then I see it's the same username. What's going on?
How do these comments "feel" AI written? This post in particular looks like something I would have written myself.
It doesn't appear AI written to me. Neither does the other post you commented on. Of course, I could be wrong, but it's not nice to just accuse people of having AI written comments, especially when it's not completely obvious that's the case.
Protip: replace "I feel like" with "I think". It forces you to ask yourself why you think the way you do. This shift in speech helped me. I'm sure it will do the same for you.
i been working on something like this for my own stuff. my drive got screenshots pdfs md files invoices and random logs and i always forget what i named stuff from years ago
what helped me was
- ran ocr on images with tesseract (slow but it works)
- used unstructured and langchain to parse and chunk stuff even spreadsheets and emails
- embedded chunks with sentence-transformers and indexed it with faiss
- then built a local llm agent (used a quantized mistral model) to rerank results smartly
its rough but works like a semantic grep for your whole disk
if you want less diy paperless-ng plus anythingllm plus a lightweight embed model could work or wait some months and someone will wrap it all in an electron app with stripe on the homepage lol
funny how much time we spend trying to find stuff we already wrote
I've shipped stuff used in parts of Latin America and Southeast Asia and one of the biggest lessons was this: latency, not bandwidth, is what kills UX.
A 10mb download over 3G is fine if you can actually start it. But when the page needs 15 round trips before first render, you're already losing the user.
We started simulating 1500ms+ RTT and packet loss by default on staging. That changed everything. Suddenly we saw how spinners made things worse, how caching saved entire flows, and how doing SSR with stale-while-revalidate wasn’t just optimization anymore. It was the only way things worked.
If your app can work on a moving train in Bangladesh, then it's gonna feel instant in SF.
theres a big gap now between interactive chatbots and fully automated code gen at build time. so a tool that integrates with the build pipeline and produces documented code would be a game changer.
also if it explains why it added code not in your specs that would improve trust a lot. this is possible with current models and some clever tooling but nobody cracked the ux yet.
personally ive tried a few hacks with rust macros and cargo steps to prototype this and it works ok but still rough around edges.
this will definitely be a hot area in next 1-2 years