01 / 03
Guest-facing AI agents
Inquiries, bookings, support — answered in seconds, in your customer's language. Trained on your business.
Like: Ranger at Mara Hilltop, below.
// AI BUILD TEAM · EMBEDDED WITH YOUR BUSINESS
Two weeks.
One working AI agent.
Live in your business.
Two of our engineers embed with yours and ship your first AI agent in two weeks.
Penn State · Ex-Microsoft · Ex-Swiggy · Top 1% Upwork · $300M+ moved through systems we built · Front page of Hacker News · 470 upvotes · Distributed across California, India, Kenya
Scroll to see three of them running in real businesses right now
14+ YRS
// Building software for other people's businesses
$300M+
// Moved through systems we built
50+
// Projects shipped, every one of them live
3 VENTURES
// Of our own businesses run on AI we built
// WHERE TO START
Most owners we talk to know AI matters. They just don't know where it fits in their business.
Asking "where can we use AI?" gets you a chatbot nobody opens. The better question: what's the work your team is buried in because the only fix you ever saw was hiring three more people? That's what AI is actually good at, and it's where we start.
// THE QUESTION WE START WITH
If twice as many customers came through your door tomorrow, where would your business start breaking? That's where we start.
// WHAT WE BUILD
Pick a bottleneck. We build the agent.
02 / 03
Ops & back-office agents
Reservations, scheduling, quoting, routing — the work that eats your team's week, run in the background.
Like: Mara Hilltop staff ops, below.
03 / 03
Autonomous decision agents
Pricing, trading, restocking — repeat calls your team makes, made faster and more consistently.
Like: Pear's Agent Pair, below.
// REAL THINGS WE'VE BUILT
Three businesses we built AI for. All three are running today.
Two we built for clients. One we built for ourselves. All three are live and you can click through and see them.
Hospitality · AI Operations
An AI agent that runs a safari lodge in the Maasai Mara.
If you've ever wondered what AI could actually do for a small business with real customers, this is what it looks like.
The problemSmall team, everything manual. Guest inquiries unanswered at 2am. Staff buried in ops instead of guest experience.
What we builtA guest-facing AI (Ranger) that handles inquiries, quotes, and bookings 24/7. A staff operations agent that manages reservations, check-ins, pricing, and reporting through conversation.
What changedHours of daily manual work gone. Inquiries answered in seconds, not days.
View project ↗R
Ranger
Mara Hilltop · online
LIVE
DeFi · Autonomous Trading
An AI agent that trades perpetuals on signals, sentiment and strategy.
Different industry, but the same shape of problem: a decision that has to be made faster than a human can make it. We built the system that makes it.
The problemTrading decisions required constant human monitoring of on-chain data, sentiment, and market conditions — more signals than any human can watch.
What we builtAn autonomous agent that analyzes signals, evaluates conditions, and manages long/short positions on perpetual contracts — independently.
What changedMulti-factor trading at machine speed, 24/7. Live on Arbitrum and HyperEVM for 2+ years.
View project ↗ Agent Pair · Signal
Sentiment divergence +2.4σ · entering pair
Dev tools · Agent-native publishing
A blog platform where the AI agent is the user.
This one we built for us. Every blog post on simbastack.com/blog and blog.marahilltop.com goes through it. We had our own bottleneck, and we built the fix in public.
The problemExisting blog tools assume a human at the keyboard — OAuth flows, draft UIs, image uploads. AI agents trying to publish hit a wall of human-shaped friction.
What we builtA publishing platform with the agent as a first-class user. Tell Claude or ChatGPT to publish a post, the agent calls the MCP tool, and a live URL comes back in seconds.
What changedZero human steps from AI prompt to live URL. Open-core, live at slopit.io.
View project ↗slopit.publish({
title: "Why remote work wins",
body: "...",
tags: ["remote", "work"]
}) → 200 OK · 142ms
Live at nj.slopit.io/why-remote-work-wins
// HOW WE ACTUALLY WORK
Forward-deployed engineers.
Not a vendor. Not an agency.
Two engineers from our team work alongside yours. They sit in your Slack, learn your business, and ship working software in your repo.
AI doesn't fail at the model. It fails at the boundary — between the engineer's assumption and the business's actual workflow. Strategy decks describe that boundary. Forward-deployed engineers stand on it.
If you wanted those engineers in-house: a six-month search, $300K-plus loaded comp each, three months lost to onboarding. Or take ours — two engineers, already a team — embedded with yours next week.
// HOW THIS WORKS
From "I'm curious" to AI running in your business. About 2 weeks.
01
A real conversation, free
45 minutes with NJ. You walk us through how your business works today, where it's slow, and what you've already tried. We tell you straight up where AI helps you and where it doesn't yet. You don't get a deck. You don't get a proposal-by-PDF.
02
Week 1: discover. Week 2: ship.
Two engineers in your Slack. Bottleneck found and build plan written by Friday. End of week 2: an agent live in your environment, doing real work. Fixed price, fixed scope, in your repo.
03
We keep going, or we don't
If it's working, we find the next piece and keep going. If something's off, we fix it before we move on. You're never stuck in a 12-month contract for a system that isn't pulling its weight.
// PRICING
Clear pricing. No discovery calls to find out what things cost.
Start here
AI Build Sprint
$12,500
Two weeks to a working AI agent, live inside your business. Fixed price, fixed scope.
- Week 1 discovery, week 2 the agent ships live in your environment
- Built inside your existing tools and data
- Team walkthrough and docs
- 30 days of post-launch support
Larger multi-system builds — multiple integrations, complex permissions, production-critical workflows — scoped separately.
If we can't find at least one high-ROI opportunity on the first call, we don't take you on.
Start a build sprint →Embedded AI Team
$10,000/month
For businesses that want to keep shipping after the first agent is live.
Two engineers embed with your team — they monitor what we built, improve it, and ship the next AI workflow every month.
Two-month minimum. Cancel anytime after.
- A new AI workflow shipped every month
- Monitoring, tuning, and fixes on everything we've built
- New integrations as your business grows
- A direct line to the engineers who build — no account managers
Start with a build sprint. Keep us embedded if the first one works.
// THE TEAM
No account managers. No layers. You work with the people who build.
Founder & Chief Architect
NJ
Penn State hackathon winner turned Silicon Valley builder. NJ has spent 14 years shipping software for other people's businesses: crypto since 2014, top 1% on Upwork, $300M+ moved through systems he built, and now AI. He also runs a safari lodge in Kenya, and he picks up the phone.
Direct line California · +1 (717) 683-9393 · call or WhatsApp
Senior Engineer & Delivery Lead
Deep
Leads client delivery end-to-end. The person who makes sure your system works in production, not just in a demo.
Senior Engineer & Systems Architect
Pranjal
ex-Swiggy. Our go-to for the hardest problems: Web3 infra, AI systems, anything where “figure it out” is the spec.
Full-Stack Engineer (Design Lead)
Chetan
Full-stack with design instincts. Most of our UI passes through him before it ships — and increasingly the AI integrations underneath it.
Full-Stack Engineer (Backend / Performance)
Hemanshu
Backend-first full-stack. Built a 250 req/sec layer behind one of our hospitality systems. Owns caching, API design, the parts that have to scale under load.
// WRITING
We share what we learn shipping production AI.
Engineering · May 2026
While I slept, my 5-year-old MacBook ran Gemma 4 locally and indexed a year of video
// LET'S BUILD
You don't need an AI strategy.
You need AI that's working.
Tell us one thing in your business you wish ran itself. We'll come back to you within a business day with an honest answer on whether AI can do it, and what that would actually look like.
Don't fill this out if you're looking for a strategy deck, a 6-month roadmap, or the cheapest chatbot money can buy.
Do fill it out if you have a specific bottleneck in your business and want an honest answer on whether AI can fix it.
Or skip the form Call or WhatsApp NJ at +1 (717) 683-9393. He actually picks up.
// Thanks — received
We'll be in touch within one business day from nj@simbastack.com.
// Couldn't send
Something went wrong on our end. Email nj@simbastack.com directly and we'll pick it up.