Describe your company. Naïve builds it.
San Francisco, CA
- Tweets 980
- Following 18
- Followers 4,083
- Likes 753
Jev (@typesafeai) is now live in Naïve.
Your agent shouldn't be using the same model for every task
Now it doesn't have to.
Read more at:
usenaive.ai/docs/concepts/mo…
7
2
21
4,751
Introducing Vetta, the most efficient harness for long-horizon agent tasks.
Same model, same tasks. Only the harness changed:
🥇 Vetta: $0.298 per finished task
🥈 claude-code: $0.872
🥉 hermes: $1.095
Highest finish rate, a third of the price. Autonomous companies only happen when the cost makes sense.
Here's how we get there:
50
50
498
408,471
We proved Vetta for ourselves first.
Old infra:
> ~$670k/mo for 750k+ agents
> ~$2M/mo running frontier claude-family inference on all tasks
> Always-on VM per tenant, billed 24/7.
New infra:
> ~$50k/mo for 1M agents
> ~$200k/mo auto routing tasks and optimising inference
> Agents sleep when idle.
1
13
2,826
Naïve Lab is live
four benchmarks on the substrate agents actually run on.
Serverless agent runtime hosting: what a million agents cost to keep alive, at 100,000 tenants × 10 agents per month.
🥇 Naïve · serverless — ~$44–60k
🥈 Self-host Hermes / Eve — ~$740k
🥉 Self-host OpenClaw / Paperclip — ~$1.5M
VM per tenant (AWS) — ~$14M
Every runtime here is free to licence. The cost is the box staying awake between turns, and agent work is bursty — a turn, then nothing for an hour.
We want an agent running continuously for every workspace. That is a hosting problem before it is a model problem.
34
24
131
405,374
Agent Brain on FictionalCharacters QA, at a matched gpt-4o-mini backbone.
Single-hop
🥇Naïve Brain 91%
🥈 CAR (prior SOTA) 78%
🥉 HippoRAG-v2 54%
Multi-hop
🥇 Naïve Brain 57%
🥈 CAR (prior SOTA) 30.2%
🥉 the remaining field ≤7%
22 published systems answer ≤7% of multi-hop questions correctly. It is the split the benchmark reports as universally failed.
1
10
528
Inference: 20–30% cheaper per solved task on SWE-bench Pro, by delegating retrieval to a scout sub-agent.
Claude Opus 4.8 — $1.23/solved, was $1.75 · −29.8% · +3 solved
Claude Opus 5 — $1.07, was $1.35 · −20.4% · +1
GPT-5.6 Sol — $0.82, was $1.05 · −21.9% · −2
Kimi K3 — $1.00, was $1.35 · −26.0% · +1
A scout call costs ~$0.0034 against ~$0.031 for a planner step — roughly 9× less and adds ~1.8% to the bill.
10
375
naïve retweeted
Coding agents can build an app in a day, building the company around it still takes months.
That gap is why I invested in @usenaive .
Naïve gives AI agents the infrastructure to run entire businesses. With one config file and one API, developers can provision payments, company formation, email, phone numbers, compute, databases, authentication, memory, model routing, orchestration, and governance.
The technical ambition is matched by the founders. Sean Dorje and Dennis Zax have been building together since they were 14. They sold their first company as teenagers, dropped out of Berkeley, joined Y Combinator, and have spent years working toward this vision.
Their early benchmarks are impressive: agents start in 2.3 milliseconds using 1.2MB of memory, while their shared context layer delivers stronger recall with 11x fewer tokens per query.
As more work moves to autonomous agents, companies will need infrastructure designed for agents from day one. Naïve is building that operating stack.
Excited to support Sean, Dennis, and the Naïve team!
22
13
158
42,217
Our AI accounting agent, Lynx, just saved us $76,510.00.
We built it fully on Codex plugged into Naive’s API which sets up all of Lynx’s tools:
> Hosted Agent Runtime: runs Hermes in an isolated microVM sandbox with cron schedules for task execution.
> 3rd party integrations: QuickBooks, Stripe and Mercury for real-time data import and sync
> Cloud deployments: built its own Postgres DB and a web-dashboard for P&L, balance sheet, cash flow, runway.
> Governance Layer: we set agent policies defined in code to ensure it doesn't make any unapproved changes and get the job done under $100/mo
Last week's reconciliation run: Lynx matched Stripe payouts and the Mercury feed against QuickBooks AR and found 23 invoices marked sent, never collected — the oldest 3 months stale.
It drafted the follow-ups, queued them in the approvals feed, and we got back $76,510.23 within 30 days.
15
3
42
31,884
Try it yourself. Or send it to your agent:
`read usenaive.ai/accounting-agent…`
$20 credit — comment "naive" + retweet
Docs: usenaive.ai/docs
1
933
I've been testing Naïve over the past few days, and it's genuinely impressive.
Really love how it consolidates almost everything you'd normally configure across multiple platforms into a single workflow.
So, you spend less time jumping between dashboards and more time actually building.
@claudeai + @usenaive >>>>
Introducing Naïve. ⛈️All-in-one infrastructure for your coding agent.
Paste one prompt → your agent writes one config file → your whole stack is live:
> Cloud, db, auth, storage (Vercel, Supabase, AWS)
> Hosted agent runtimes (Hermes, OpenClaw)
> Payments, identity, incorporation (card, KYC/KYB, email, phone, LLC/EIN)
> 300+ models, 10,000+ tools (search, video, domains, trading, browser, mobile)
> Autonomous signup via secure credential vault
Stop setting up. Just build bro.
Reply "Naive" + Retweet. Get $20 credit for free.
3
3
16
896
naïve retweeted
I challenged myself to build a real AI SaaS using just "one prompt".
No infrastructure.
No backend setup.
No deployment.
I described what I wanted, and @usenaive handled the rest—from cloud and database to authentication and payments.
The result? A live app in minutes.
This is one of the easiest ways I've seen to turn an idea into a working product.
What would you build first?
6
5
12
19,615
naïve retweeted
I've built products that worked.
the hard part was never writing the code.
it was everything that came after.
buying a domain.
connecting cloud services.
setting up authentication.
configuring databases.
creating a company.
activating payments.
verifying identity.
handling compliance.
days of setup before a single customer could even use what i built.
that's the bottleneck most builders never talk about.
what caught my attention about @usenaive is that it aims to remove that entire layer of friction.
one prompt.
one configuration.
an entire production stack ready to go.
cloud, databases, authentication, storage, payments, deployment, business operations, and the tools your app needs—all connected instead of stitched together manually.
the goal isn't to replace builders.
it's to eliminate the repetitive work that slows builders down.
less time configuring.
more time building.
less infrastructure.
more innovation.
the future belongs to people who can turn an idea into a real product faster than everyone else.
that's the shift i'm excited to see.
@usenaive is building for that future. 🚀
Made with AI
Paid partnership (ad)
5
1
12
1,224