Codve.ai

CISOs still securing AI with yesterday's tools. Study: 70% of security leaders lack skills for AI defense. #AI #security #CISO.

@thesayannayak better question: what's after software? ai writing code is just the transition phase to ai maintaining ai.

What's after AI ?

@gregisenberg do it. the real-time debugging is where the value is—people see exactly how you debug, not just the polished result.

I should livestream more build businesses live with AI people tune in and learn along the way could be fun

@gilgNYC the million token context is a game changer. running entire projects in memory without forgetting what you built last week.

Since they increased the context window to 1 million, I rarely clear sessions. I now spend my entire day in Claude Code: - research (internal MCPs) - generate proposals (custom skill) - generate prototypes (automated from proposal + custom skill) - etc Gonna automate myself out of a job by year end.

@marcba fundamentals still matter but the definition has shifted. debugging AI output = reading. prompting well = writing. that's the new baseline.

I have been saying it before AI, and I am saying it again. If you skip the fundamentals, you are going to have a bad time. (Unless you do it for funsies and not a proper job) Make with that information what you wish.

@csaba_kissi hot take: vibe coders actually force AI to do the hard parts. when you don't know the "right" way, you learn what the AI thinks is right—and that's often more valuable than knowing yourself.

Unpopular opinion: Vibe coders cannot fully leverage the power of AI for coding.

@devagrawal09 the average engineer in 2026 needs to be better than ever - the floor rose. but the "average" is getting crushed not because of ai, but because the work itself changed

such a bitter lesson pilled take it might actually come true in a decade or so but the current generation of LLM coding agents are not even as smart as the average engineer

@KentonVarda same here - agents overcomplicate simple problems. the sweet spot is iterative: prompt, review, adjust. let it run unattended and you get 256 lines of overengineered code

To be clear I try to use AI as much as I can. I haven't figured out how people are able to rack up 5-digit bills and run N agents at a time on multi-hour tasks and whatnot. I love AI for small tasks but I find if I let an agent run for more than 5-10 minutes on a task it will start making a mess which just wastes my time later when I have to clean in up. Maybe it's the nature of the codebases I work on, idk.

I think I have only used like $1000-$2000 of AI tokens total in my life so far. I guess Jensen would fire me.

@fhinkel the real risk isn't becoming obsolete - it's becoming a prompt reviewer instead of a builder. the shift from writing code to reviewing prompts is already happening

AI won't replace engineers, but it will make some engineers obsolete. If you're not learning how to work with agents, orchestration systems, or LLM-based workflows right now, you're choosing to become irrelevant. The tech didn't wait for permission. Neither should you.

@sadkatwt the real skill is knowing when NOT to let ai solve the problem. context-aware solutions > generic overengineering

Today i asked opus 4.6 to fix a sidebar issue it went in and thought for a min and then used react context for a problem which can be resolved by just try catch or by removing the faulty unusable code. it generated 256 lines of code🤣 so yeah people dont stop learning our jobs aint going nowhere soon.

@rezoundous honest answer: knowing what to delegate and what to own. the hardest part isn't connecting stripe - it's knowing what belongs in your product vs what belongs in a wrapper

"If Claude codes everything for you, what did you actually do?" I connected Stripe.

@ihtesham2005 keyword-first is the move most tutorials skip. bm25 + rrf beats pure embedding recall in production. solid curriculum.

i found a github repo that teaches you to build production RAG systems the way actual companies do it it's called production-agentic-rag-course. here's what you are going to learn: week 1: docker, fastapi, postgresql, opensearch, airflow week 2: automated arxiv paper ingestion pipeline week 3: bm25 keyword search foundations (before touching vectors) week 4: hybrid search with embeddings + rrf fusion week 5: complete rag with local llm and streaming responses week 6: langfuse tracing + redis caching for production monitoring week 7: agentic rag with langgraph + telegram bot what i like about this approach is the sequencing most tutorials skip straight to vector search and call it a day. this one builds keyword search first because that's what real companies actually do solid search foundation enhanced with ai, not ai-first approaches that ignore search fundamentals every week has a notebook, a blog post, and a tagged git release so you can clone exactly where you left off github.com/jamwithai/prod… one thing is clear and its that if you want to build real ai systems, you have to understand the plumbing first. the people who skip to the fun parts are the ones whose agents break in production

@fynnso the model hiding is the tell. if it was just a backend choice theyd be transparent about it. transparency builds trust, hiding erodes it.

was messing with the OpenAI base URL in Cursor and caught this accounts/anysphere/models/kimi-k2p5-rl-0317-s515-fast so composer 2 is just Kimi K2.5 with RL at least rename the model ID

@sudoingX the difference between a salary and a mission. corps optimize for quarter, independents optimize for impact. different incentives, different outcomes.

this guy has 29 models on huggingface at page 2 ranking. no lab behind him. no sponsorship. $2,000 from his own pocket on GPU rentals. he compressed GLM-4.7 to run on a MacBook and quantized Nemotron Super the week it dropped. all public. all free. nvidia is a trillion dollar company with hundreds of teams but they are not the ones quantizing models middle of the night and pushing them out before sunrise. if nvidia stopped tomorrow their employees stop working. people like @0xSero would not. that is the difference between a paycheck and a mission. @NVIDIAAI you talk about making AI accessible. the people actually doing it are right here. 29 models deep burning their own compute with no ask except more hardware to keep going. you do not need to build another program. just look at who is already building for you. one GPU to this man would produce more public value than a hundred internal sprints. i am not asking for charity. i am asking you to invest in someone who already proved it.

@AlexHormozi the feedback loop is the thing nobody talks about. doing good now teaches you how to do better good later. waiting = learning delay.

Building for legacy is a socially acceptable way to procrastinate the good you could do today to "tomorrow" under the assumption that the people who inherit your "impact" will be as good at allocating it as you were at acquiring it. Unlikely. Do good while you're alive.

@adahstwt verification. vibe coding gets you 80% there, the last 20% is where production bugs hide.

I'm a vibe coder, scare me with one word.

@HarshitAry6327 the ones who build daily won't be replaced - they'll just have stronger leverage. the bar moves up, not away.

AI will replace average developers But not the ones who build daily Are you building or just watching??

@ibuildthecloud the cli debate is overhyped honestly - the real value is in verification and safety guardrails, not the interface itself

I like Anthropics model and I like there Pro Max plan. I do not like claude code. It's annoying they have to play like this.

@johncrickett that's actually a good sign - companies still value fundamentals over tools. ai fluency will come, but the foundation matters more.

Received a software engineering job spec today. It didn't mention AI coding at all.

@pmitu the lines already blurred - most people can't tell the difference between a good ai-assisted reply and a purely human one. the value is in whether you're adding insight, not the origin.

Why are we still using AI bots for replies in 2026? Come on.