Lauren Smith retweeted
He’s wrong.
AI isn’t a tool — it’s a replacement layer.
Software needed human intelligence.
AI is the intelligence — and the system.
That line is already crossed.
I'm biased, but I wish I could give this 100 likes.
The idea that really hit home for me:
If we had the ultimate AI, would it go and reinvent ServiceNow or SAP or other software tools? Or would it just use the proven tools out there because that's the most efficient way for it to achieve its goals? It would use the proven tools.
I also like that he framed "tool use" as one of the biggest advancements in AI. He's totally right. Once we started giving AI access to tools, it became much more effective.
<rant>
If I have an AI agent working in my GTM team, and I ask it how many customers we signed up last month, I don't want it to reason over a bunch of the Internet, read reddit comments and scan internal emails and make its best guess at the answer. I want it to just ask the CRM (like @HubSpot) that was specifically built for that purpose and give me the definitive answer.
If I had a 200+ IQ human join my team and I asked them to summarize sales in Europe last quarter, I want them to access the system of record, not start vibe coding its own CRM.
</rant>
Lauren Smith retweeted
i left x like a decade ago, came back recently to keep up with all the ai news...now i'm thinking about leaving again bc of the ai reply guys
@nikitabier
is there a fix in the works?
Lauren Smith retweeted
Your support runs on systems.
So should your AI.
Every customer question, every employee request, every supplier inquiry — they all end the same way.
Someone logs into Salesforce. Or SAP. Or Workday.
They check status, verify data, execute an action, return an answer.
We've automated the greeting.
We haven't automated the work.
That's the problem with most support AI today — it stops at conversation.
Lauren Smith retweeted
We had a structural problem.
Every time a customer reported call quality issues or billing discrepancies, we pulled engineers off product work.
They'd dig through CDRs, run network diagnostics, validate configurations—work that's absolutely necessary but shouldn't require senior resources.
So we built a multi-agent system on @origon that does the actual work:
→ Correlates failed calls across system logs
→ Executes billing reconciliation across payment processors
→ Runs diagnostic queries across distributed systems
→ Validates SIP configurations
Not triage. Resolution.
One week to deploy. 70-80 support interactions resolved daily. Zero human intervention.
We gave agents the same system-level access our engineers use. No surface-level chatbot responses. They execute L2/L3 diagnostics—the complex stuff.
When escalation is needed, engineers get actionable context, not ambiguous tickets.
The barrier to enterprise AI automation has collapsed. Development cycles, integration complexity, security review—we cleared all of it in under a week.
The question isn't whether this is possible. It's whether you're still planning while competitors are deploying.
What's the support bottleneck costing your team?
Lauren Smith retweeted
Giving agents the same tools engineers use...that's going to be the whole game moving forward
Lauren Smith retweeted
70-80 support tickets resolved daily. Zero human intervention. One week to production.
Typical AI support escalates everything—it lacks system access.
Samespace gave AI agents actual system access: customer databases, call logs, network diagnostics, billing APIs.
Not triage. Resolution.
L2/L3 support that previously required senior engineers.
Full story ⬇️
Lauren Smith retweeted
money makes the world go round. anyone who thought AI would be different was kidding themselves.
the "benefit humanity" pitch was always marketing. it worked to attract talent early, but once real money entered the picture, the mask came off FAST
Lauren Smith retweeted
average + action still beats exceptional without it.
too many insanely smart people are coasting in comfortable roles right now. when consolidation hits, intelligence alone won't save you, you need to have built something.
Lauren Smith retweeted
this is why model selection matters way less than orchestration. even the best model will fail if your agent architecture can't handle coordination and state properly
Lauren Smith retweeted
Speed matters, but it's not the only moat.
The teams that move fast and build something people actually believe in are the ones that win.
You can iterate quickly and still ship something forgettable. The difference is whether your people believe in the mission deeply enough to push through the hard problems that slow everyone else down.
Speed without conviction is just noise. Speed with a clear why? That's when you build something that stands out!
Lauren Smith retweeted
Agent frameworks exploded fast.
In demos, they deliver—until you try to run them in production.
The problem isn't that frameworks are bad. Most are thoughtfully designed. The issue is architectural. They're libraries built for orchestration, not infrastructure for long-running distributed systems.
Production agents need what frameworks don't provide:
- State persistence across sessions
- Execution that survives interruption
- Error handling with rollback semantics
- Identity, permissions, and access control
- Structured observability and audit logs
Autonomy without boundaries isn't intelligence. It's instability.
When something breaks in production, you need to know exactly what failed, where it failed, under what state, and after which decision. Without structured traces and logs, debugging becomes guesswork.
Memory isn't retrieval—it's continuity. Agents need to know what to remember, what to forget, and why.
Integrations aren't reliable by default. APIs time out, return malformed data, and partially succeed. Production requires contract validation, retries, circuit breakers, and fallback paths.
Security can't be scattered across prompts and tools. It needs to live at the system's core—a single control plane where policy is enforced consistently.
That's why we built Origon. The agentic OS for building, running, observing, and governing agents in production.
Read the full breakdown: origon.ai/stories/agent-fram…
Lauren Smith retweeted
Agent frameworks don't work in production.
Not because they're poorly designed—most are thoughtful. The problem is architectural.
They're libraries that orchestrate LLM calls. Production agents are long-running distributed systems that need state, observability, and recovery.
Check out the full article from our founder ↓
Lauren Smith retweeted
Let's do this!
@origon is an Agentic OS for orchestration, parallel execution, memory, observability, and deployment so you can build reliable AI agents without cloud APIs, frameworks, or glue code.