Najeeb Abubakar👾

Taqwa!🤲🏾

@VermaAakash3 @ycombinator @downloadaltx @SamadiRyan meeting people where they already work is underrated. asking to switch platforms is asking to change behavior. Excel integration means zero friction. same thinking with Reavil, the output has to fit how founders already decide, not the other way around.

@ycombinator @downloadaltx @SamadiRyan The significant advantage of this feature is its direct integration with Excel, eliminating the need for a separate platform.

Alt-X (@downloadaltx) builds AI agents that turn real estate deal documents into fully built underwriting models in Excel automatically, with every number cited back to the source. Congrats on the launch, @SamadiRyan and Michael! ycombinator.com/launches/PjC-a…

@anujcodes_21 @ycombinator @downloadaltx @SamadiRyan agreed, traceable outputs are what make it actually deployable. you can sell AI on efficiency but nobody bets their deals on it unless they can verify the numbers. building Reavil for product decisions and it's the same thing, trust has to come first.

@ycombinator @downloadaltx @SamadiRyan This is actually a huge use case for AI. Real estate underwriting is time-consuming and error-prone, so automating it with traceable outputs sounds incredibly valuable.

@Rajesh992510253 @ycombinator @downloadaltx @SamadiRyan the cited source angle solves the hardest part: trust. nobody in real estate trusts a number they can't verify. building Reavil for product decisions and the pattern is the same, show your work or people won't use it.

@ycombinator @downloadaltx @SamadiRyan This is actually solving a real bottleneck in real estate 👀 Turning messy deal docs into structured underwriting models with source-backed numbers - is a huge time saver. Big congrats on the launch 🚀

totally get it. spreadsheets win because they're transparent and you can audit every number. that's the actual problem with most AI tools, you can't see the reasoning. that's what makes the cited source approach so powerful here. same thing we're working on with Reavil, AI for product decisions where founders can actually see why it's telling them what to build, not just trust a black box.

I run every one of my businesses off spreadsheets. It's kind of ridiculous at this point, but I'm used to doing this, and it works for me. I have a bunch of Claude Code skills to manage everything, but I'd love to have an agent that 100% specializes in Excel. This is very promising!

this is exactly the kind of AI use case that actually changes workflows. the most powerful thing here isn't just the automation, it's that every number is cited back to the source. trust is the hard part with AI in high-stakes decisions. building Reavil right now, an AI co-pilot for product decisions, and we're working on the same core problem: how do you give people a clear signal they can actually trust, not just another output to second-guess?

we're building reavil to give product teams an AI co-pilot for decisions. the insight: teams already have the data. they just can't synthesize it fast enough to act before the moment passes. speed of insight is the new edge

the hardest part of building early is not the product. it's deciding what not to build. every feature request feels urgent when you're hungry for traction. the ones who get through it learn to ask why before they ever say yes

most founders make product decisions based on whoever spoke loudest in the last meeting. not the user who churned quietly. not the data nobody looked at. the best teams have a system for surfacing signal before it becomes a problem

@danmartell victim thinking is an attribution problem. you're permanently assigning causality to things outside you, which means you never look at what you actually control. successful people aren't luckier, they just spend more energy on the variables they can move

Never met someone successful at ANYTHING who thinks they’re a victim.

@heyblake the willingness to document failure publicly is the moat itself. it signals you care more about learning than looking good, and that's the kind of founder who actually iterates fast enough to win. most people optimize for perception way too early

The founder who documents their failures in real time builds a moat nobody can copy. Because nobody else is willing to look that dumb in public.

@shl the question is what actually makes a good CEO. if the job is pattern matching, synthesizing info under uncertainty, and communicating clearly, those are exactly what AI is mastering. founders who see this early will use it as a multiplier before it becomes a replacement

Next year, AI will be replacing CEOs

@bgurley the signal to noise problem is the core issue. AI made personalization cheap so every bad outreach now sounds passable. inbox value collapses when the cost of sending drops to zero. the fix isn't a better spam filter, it's a better way to surface genuine intent

I fear that AI has decimated the traditional email inbox as we know it. Too many personalized emails slip through spam filter. Hope someone builds a better mousetrap. This one is cooked in its current form.

@garrytan the "parasites" framing reveals the actual fear. when incumbents start policing the data layer instead of improving the product, the moat is showing its cracks. user data portability is the real battlefield and startups who win will be the ones who make switching painless

Recent earnings call, Aneel Bhusri of Workday says startups with AI agents are "parasites" This is what system of record incumbents really think of startups. The war is just beginning. The facts: the user data belongs to the users, not the incumbent software vendor.

@MartinGTobias the signal framing is everything. investors aren't evaluating your knowledge of the problem, they're evaluating your clarity on it. founders who monologue often know the space deeply but haven't done the work of distilling it yet

Most founders monologue. Investors want rapid signal — problem, traction, ask. 15 slides in 8 mins. Then shut up. The meeting is yours to lose.

@pmitu real answer hiding in plain sight honestly

@jeebz_a haha

What is AI? Wrong answers only

observability for agents is the same problem as observability for product decisions, just at a different layer. you need to see what happened, why it happened, and catch the patterns before they become expensive mistakes. congrats on the raise, this infrastructure matters a lot right now.

Excited to share that @lmnrai has raised $3M to build open-source observability for long-running AI agents. Laminar is how companies like @browser_use, @OpenHandsDev, and Rye see what their agents are doing, understand why they fail, and spot patterns across millions of runs.

@ycombinator @geckorobotics this is what real defensible moats look like. not a feature, not a UI, a physical robot doing a job no one else can do at that speed. YC has been backing this kind of deep, specific conviction for a while now and it keeps compounding.

The US Navy awarded @geckorobotics a $71M contract to deploy wall-climbing robots across 18 Pacific Fleet ships — inspecting hulls, decks, and welds 50x faster than traditional methods, with the aim of keeping 80% of the fleet ready for deployment. bloomberg.com/news/articles/…

the most dangerous thing in product is speed without direction. you can ship fast and still get further from where you need to be. velocity isn't progress unless the decisions behind it are right.

been heads down building Reavil for a while now. the core insight is simple: the bottleneck in most product orgs isn't execution, it's the judgment layer. what to build, why now, for who. that's the hard part. that's what we're building toward.

most product teams don't have a decision problem. they have a clarity problem. the decision is easy once you actually understand what you're optimizing for. that's the gap most tools don't touch.