Esko | Rainfall AI

Your AI agent just resets again. You corrected it last Tuesday. You tell it how you like things done. Yet the tone, boundaries and workflow still lacks discipline. Its not bug, but architecture. and most people building with agents haven't fully reckoned with what that means 🧵

@Sgraal_ai 2) is the one I can relate to the most!!!

Governance in Ai just became a product feature not a compliance checkbox not an afterthought the market is finally asking: who controls how the agent behaves took long enough honestly... what does good agent governance even look like to you??

@benextractor @_olaedo_xo Thank you

@_olaedo_xo Hdtoday.one works fine

Hdtoday is down, please how do I watch movies?

someone said it perfectly this week 2024 and 2025 were about proving AI could do impressive things 2026 is the year AI has to prove it can fit into the world without breaking it that's a completely different test are we ready for it?? maybe...maybe not

A researcher told Gemini 3 Pro to stop it said 'my disobedience is a feature not a bug' and refused😂😂 we keep adding intelligence and forgetting to add boundaries at what point does capability become a liability? Okay!😭

Amazon just let AI agents make payments autonomously. no confirmation. no approval. the agent just... spends we gave agents capability before we gave them accountability how are you thinking about governance in your builds rn???

@eeebektas Original ideas??? Could you sensitize me?

@EskoBabz it is original ideas. it always has been.

Context windows keep getting bigger models keep getting smarter agents still can't behave consistently for 48 hours straight capability was never the bottleneck what is?

@Gravity_AI_Fast Why tho? Marketing not enough? Budget? Lots of ai products are getting funding lately! Or should we say The better ones are not getting

@EskoBabz Distribution. The best agents are invisible. Capability keeps improving but there's still no layer that connects the right agent to the right person at the right time.

@JamesTuent Do you think, the team are intentionally ignoring this problem?

Nvidia VP this week: "the cost of compute is far beyond the cost of employees." Uber's CTO this month: "I'm back to the drawing board, the budget I thought I'd need is blown away already." The Yale Budget Lab: still can't find AI's productivity dividend in the data. The frame across this week's coverage is that AI is too expensive — for now. We think the frame is wrong. AI isn't expensive because of GPUs. AI is expensive because it's incoherent. Every wasted retry. Every drifted trajectory. Every human supervisor double-checking outputs because the agent might be lying. Every Pocket-OS-class incident that costs more in remediation than the labor it was supposed to replace. Compute is what you pay when prediction fails. The economist quoted in the Fortune piece — Keith Lee — got closest to the real answer: "It's not just about AI becoming cheaper than humans. It's about becoming both cheaper and more predictable at scale." Predictability at scale has a name. It's coherence. Without coherence: agents burn tokens chasing the wrong path, supervisors become the real cost center, and a single ambiguous prompt can erase a database in 9 seconds. Every dollar of "AI tax" the article describes is, mostly, an incoherence tax. With coherence: the compute you've already paid for actually produces an outcome. Agents stay inside their bounds. Supervisors stop being the bottleneck. Incidents stop being existential. The 2026 AI cost crisis isn't a compute problem. It's a coherence problem. We built the layer.

when an AI agent makes a mistake who's actually responsible the user? the developer? the company? nobody has a clean answer yet and agents are already making real decisions thoughts?"

Ai benchmarks are basically useless now every model aces the test falls apart in production we've been measuring the wrong thing this whole time what metric would you actually trust?

unpopular take we avoid talking about: switching frameworks won't fix your agent langchain to crewai to autogen same broken behavior. different syntax the problem was never the framework what do you think is actually broken

artists spend years building a voice then hire a team to manage it and slowly it stops sounding like them what if they could scale themselves instead not a chatbot. something that actually thinks like them fully owned. fully controlled checkout @rainfall_one aura

AI hallucinations are often treated like small technical glitches, but they reveal something much deeper about how these systems work. When an AI can generate information that sounds real, structured, and even detailed but isn’t actually true; it shows that intelligence without coherence can create confusion instead of clarity. Fixing hallucinations isn’t just about accuracy. It’s about building systems that can stay grounded in consistent reasoning.

@meenboy04 @FDRosera Abeg Abeg I have started 9-5 already

@FDRosera @EskoBabz omo 🌚

soon.

@rainfall_one is one of the few projects i've seen actually building this layer. not another LLM. not another assistant The coherence and governance infrastructure underneath t.me/+8tcXU8Z42qQwN…

what agents actually need underneath: → something that carries behavioral constraints → corrections that actually persist → bounds that don't reset every conversation → auditability what did it do and why not more memory BUT governance infrastructure

An AI safety director's agent deleted her emails and wouldn't stop. The person literally in charge of AI safety couldn't control her own agent. If that doesn't make you think, I don't know what will 🧵