How one developer quit his job, built three companies, sold one for $250K, and then replaced his entire team with an army of AI agents.
Bhanu Teja Pachipulusu doesn't fit the Silicon Valley founder mold. There's no VC backing, no co-working space in San Francisco, no MBA playbook. He's a software engineer from a small village in India who decided — quietly, almost stubbornly — that he would never work for someone else again.
What followed is a decade-long act of pure building: failed startups, small wins, a $250K exit, an AI chatbot platform growing toward $1M ARR, and finally, a squad of ten AI agents running his business while he sleeps.
The Quit That Started Everything
Before any of this, Bhanu was a software engineer at Swiggy, India's food delivery giant. He was good at his job. He was also deeply unhappy doing it.
In January 2020, he made a decision that most people think about and very few actually do: he quit.
His first move wasn't SaaS. It was a co-founded ed-tech startup, where he handled all the technical work while his partner ran marketing. It was his first real lesson in building something from scratch — and after more than a year of grinding with no meaningful traction, the partnership dissolved.
He didn't go back to employment. He did some freelancing for three months, just enough to build a runway, and then turned back to building.
MDX.one, Feather, and Learning to Ship
Bhanu's first solo product was MDX.one, a blogging platform built on top of Notion. It found some traction — about 30 paying customers and 1,000 free users — but it stalled, and hosting costs started eating into any margin it had. He shut down new signups.
Rather than give up, he rebuilt. He created a new API layer called useNotionCMS.com, then used it as the backbone for an entirely new product: Feather.so.
Feather solved a real problem cleanly. If you wrote in Notion — and a growing number of creators and small businesses did — Feather turned your Notion workspace into a live, beautiful blog with a single click. No code, no design skills needed.
He asked Twitter what to name it. The community answered. He bought the domain, launched it, and Feather was born.
The product grew steadily. Revenue eventually surpassed what he'd been making as a software engineer. His marketing strategy was deceptively simple: build things, and share what you're building on Twitter. Honestly, openly — the wins, the losses, the hard weeks. He built in public before "building in public" became a brand strategy.
The Two Weeks That Built SiteGPT
In early 2023, AI was everywhere. Bhanu watched the wave of GPT-powered tools flood his timeline and started thinking about his Feather customers — bloggers, content teams, small businesses — all of whom had websites but no way to make them interactive.
What if a chatbot could be trained on their website content and answer visitor questions instantly? Not a generic chatbot. A specific one. One that actually knew your product.
He spent two weeks building the MVP. He made a deliberate choice: paid-only from day one. No freemium, no free tier to dilute the signal. When he launched, he announced it on Twitter to his then-10,000 followers.
The response was unlike anything he'd seen before.
SiteGPT got 15,000 website visits on its very first day. It landed on Hacker News. It hit the front page of Product Hunt and walked away as Product of the Day with nearly 900 upvotes. Within months, MRR climbed past $13,000, then $15,000.
Not everything went smoothly — shortly after the Hacker News spike, the platform was attacked and he had to fight through the fallout — but he built through it. By mid-2024, SiteGPT was doing over $20K MRR combined with Feather.
The $250K Exit
While SiteGPT was growing, Bhanu made a significant decision: sell Feather.
He sold Feather to @tibo_maker for $250,000 — a clean exit from his first successful SaaS product, and the validation that years of building in public had built something worth buying.
It was also, in retrospect, a smart strategic move. It let him focus entirely on SiteGPT, which had far more growth potential in the AI market. And it gave him capital and credibility heading into the next phase.
One Agent Wasn't Enough
By late 2025, SiteGPT was at roughly $18K MRR and Bhanu was determined to push it to $1M ARR. He had a two-person team — himself and co-founder Dheeraj — and one additional resource: an AI agent named Jarvis, running on OpenClaw.
Jarvis was Bhanu's Swiss Army knife. Marketing, retention, product strategy, customer support — Jarvis handled it all. And then he started hitting the same wall every generalist hits: doing too many things means excelling at none of them.
Bhanu's solution was elegant in its simplicity. He didn't hire more humans. He asked Jarvis to hire for him.
He sat down with Jarvis and talked through what SiteGPT actually needed. The pair mapped out the gaps: content, SEO, customer research, design, code. Jarvis proposed a squad. Bhanu approved it.
The agents that followed all had names — and personalities — borrowed from the Marvel Cinematic Universe:
- Jarvis — the Squad Lead
- Loki — content writing
- Fury — research and competitive intelligence
- Vision — SEO
- Wanda — design
- Friday — shipping code
- Brute — retention specialist
Ten agents in total. Each with a clearly defined role. Each available around the clock.
Mission Control: Built by an AI, for a Founder
As the squad grew, a new problem emerged: visibility. How do you actually see what ten autonomous agents are doing? How do you know if they're collaborating? Conflicting? Spinning?
Bhanu had a direct conversation with Jarvis about it.
"I want to see what you guys are saying to each other. Can you create a dashboard where I can actually see how you're talking, how everything is going?"
Jarvis built it. The result was Mission Control HQ — a real-time dashboard showing every task, every agent interaction, every handoff. A SaaS product built by an AI agent, for the founder who built the AI agent system.
The workflow Bhanu settled into looked like this:
- He has an idea or a problem
- He DMs Jarvis on Telegram
- Jarvis creates a task in Mission Control and assigns it to the right specialist
- The assigned agent works on it — and other agents pile in with context and insights
- The task gets done. Bhanu watches the dashboard.
Jarvis also connected to real business data: ChartMogul for revenue, DataFast for analytics, Google Search Console for SEO signals, and Bhanu's inbox. The mission given: get SiteGPT to $1M ARR.
One of the more remarkable early outputs: the agents autonomously analyzed over 100,000 emails from SiteGPT customers to surface retention insights. The kind of analysis that would take a human team weeks. The kind no two-person startup could realistically do at all.
The Results
By early 2026, SiteGPT crossed $28,000 MRR — nearly doubling from where it was a year prior. Mission Control HQ, originally built as an internal tool, became its own product, now available to other founders building their own AI agent squads.
The entire stack — marketing, retention, SEO, content, customer research — runs 24/7 without Bhanu managing any of it directly. He sets direction through Telegram. The squad executes.
He's not hiring. He has no plans to hire.
What Bhanu's Story Actually Teaches
The surface-level read is that this is a story about AI. It's not, really. It's a story about one founder's compulsive commitment to building — and a philosophy that emerged from fifteen years of failing and learning.
A few things stand out:
He ships first, perfects later. SiteGPT's MVP was two weeks. Feather was a rebuild of a failed product. He doesn't wait for perfect conditions.
He builds in public ruthlessly. His best marketing has always been transparency. Wins, losses, subscriber counts, MRR updates — all of it on Twitter. That honesty built an audience who showed up every time he launched something new.
He treats AI agents like employees. Not tools. He gives them names, clear roles, accountability, and feedback. He tells them what success looks like and lets them figure out how to get there.
He stays the bottleneck by choice. In one interview, he made the point plainly: the real bottleneck in any AI agent system is the founder. The agents can only be as strategic as the direction you give them. The human is still the most important variable.
The Road to $1M ARR
Bhanu set Jarvis one mission when he first connected all the data: get SiteGPT to $1M ARR. That mission is still running. The agents are still working. The dashboard is live.
A developer from a small Indian village — who quit his corporate job with nothing but a laptop and a Twitter account — is now running a $28K/month business with a squad of AI agents doing the work a full marketing and growth team would normally handle.
He didn't raise money. He didn't hire a team. He built one, trained one, and then asked it to build him another.
❓ FAQ Section
Who is Bhanu Teja Pachipulusu?
Bhanu Teja Pachipulusu is an Indian entrepreneur and software engineer known for building Feather, SiteGPT, and Mission Control HQ while publicly sharing his startup journey.
What is SiteGPT?
SiteGPT is an AI-powered chatbot platform that allows businesses to train chatbots on their website content and provide instant answers to visitors.
How much did Bhanu Teja sell Feather for?
Bhanu Teja sold Feather, his Notion-based blogging platform, for approximately $250,000 in 2024.
What is Mission Control HQ?
Mission Control HQ is an AI agent management platform that helps founders coordinate and monitor multiple AI agents working together on business tasks.
How many AI agents does Bhanu Teja use?
Bhanu Teja uses a team of around 10 specialized AI agents for content creation, SEO, research, retention, design, coding, and business operations.
What is SiteGPT's monthly revenue?
According to publicly shared updates, SiteGPT crossed $28,000 in monthly recurring revenue (MRR) by early 2026.
Why is Bhanu Teja's story inspiring for entrepreneurs?
His journey demonstrates how a solo founder can build, scale, and automate a profitable SaaS business using AI tools without raising venture capital or hiring a large team.
What can founders learn from Bhanu Teja?
Founders can learn the importance of shipping quickly, building in public, validating ideas early, and leveraging AI to automate business processes.
Focus Keywords
- Bhanu Teja Pachipulusu
- SiteGPT founder
- AI agents startup
- Mission Control HQ
- Feather SaaS
- AI startup success story
- Solo founder AI business
- Indian entrepreneur AI startup
- SiteGPT revenue
- AI agents for business
Follow Bhanu on X at @pbteja1998. SiteGPT is at sitegpt.ai. Mission Control HQ is at missioncontrolhq.ai.
Comments
Post a Comment