Does AI-assisted writing count as plagiarism? International student and future aerospace engineer Boulama Kandine doesn’t think so. He believes it jogs students’ brains in times of writer’s block and helps them write better papers. That’s why he runs his own $84,000 ARR AI writing tool for university students.
While in college, Boulama created his AI writing SaaS, Tinq.ai, at the request of clients of his Niger-based software development agency, Otimbi Labs. His clients wanted a customizable, automated system for summarizing and sorting emails, but no such app existed.
After successfully launching Tinq to complete both tasks in late 2021, Boulama searched for new markets for his product. Inspiration struck when he decided to target his application to his fellow students.
While studying, Boulama noticed classmates already used software services like Chegg for plagiarism checking and preparing for exams. He realized his email paraphrasing tool would also help them write research papers. Better yet, there was little competition among writing aid startups for the student market. AI writing apps like Jasper mainly targeted salaried professionals rather than undergraduates strapped for cash.
To make Tinq more desirable for students, Boulama added a plagiarism checker then posted about it in forums and beta-tested it on his classmates. By late 2021, he’d created a low-churn product bringing in a healthy $7,000 in MRR with a two-person team.
Today, the tools on Tinq.ai are used and whitelabeled by hundreds of small businesses, clients, and students alike. Boulama attributes all of his success to personally emailing, calling, and meeting with customers to better understand his market. Here’s how he did it.
From Minecraft to Aerospace Engineering
Boulama learned coding at fifteen to modify the popular sandbox video game, Minecraft. Later that year, a childhood friend explained to young Boulama that he could use coding for more than video games – websites, for instance.
That summer, as a fifteen-year-old, Boulama became hooked on developing online tools. Using his father’s office internet, he built free websites for anyone in his city who dropped him an email. After a year of portfolio-building, he started charging for his website service, creating a reliable side income throughout high school.
As you might have guessed by now, Boulama is a smart guy. When he graduated high school in 2015, he moved from Niger to the USA to improve his English because he wanted to study aerospace engineering there. After one year, he put his studies on hold in 2016 and took a short vacation back to Niger to begin his first startup.
While freelancing, Boulama connected with other developers in Niger and abroad for help on large projects. By 2016, he’d amassed a stacked Rolodex of skilled developers he worked with almost daily. That year, Boulama turned that list of developers into an agency called Otimbi Labs. He hired and subcontracted everyone in his network who was interested. “I know firsthand how hard it is for a dev to build a portfolio in the real world, and I hoped it would help them kick start their careers,” he says.
Returning to the US in 2017, Boulama studied engineering in California for three years then enrolled at the Florida Institute of Technology (FIT) in 2021 – as close to NASA as possible – bringing his startup with him. Once established, he registered Otimbi Labs in the State of Delaware through a Niger-based holding company and became a full-time student and business owner.
As Boulama attended undergraduate classes, Otimbi Labs became an increasingly specialized agency. By 2020, the now 30-person team had carved out a small name for itself by designing enterprise resource planning (ERP) software for small-to-medium sized businesses. These were complex projects – Boulama says one example of their work was a real-time car tracking system for a large Southeast Asian car rental company.
In 2020, Boulama’s clients at Otimbi labs began asking for word-processing automation for email replies. Boulama’s clients didn’t like other automation services on the market because they weren’t customizable.
“Businesses receive many emails in one inbox and then have to sort them manually to know where to send them,” says Boulama. “For example, technical emails go to Jim, and ops emails go to Mary. My clients wanted a system that could do that automatically.”
Boulama used open-source language models like Google Pegasus to develop a limited email-sorting tool as well as a paraphrasing-and-summarizing tool (for better record keeping) and launched it to his clients in November of 2021. They loved it. Then, Boulama started thinking about how he could sell it to more audiences.
“I would say the best [marketing] strategy, just like in any business, is word-of-mouth.”
He attempted paid advertising with a small ($100) search campaign on Google to find more clients, but it was unsuccessful. He hasn’t spent a dime on paid ads since. Shortly after, Boulama created a much more fruitful affiliate program offering 20 percent of the proceeds from referred subscribers for referers’ first two years.
“There was one particular affiliate that was relatively well-known in his niche that brought around fifty people in one month,” says Boulama. “Once people signed up through affiliates, I would send regular emails offering them a free month with the tool.”
Word Processors for Students
Despite his successful affiliate program, Boulama wanted to build a larger audience. He began seeing his fellow college students at FIT as potential customers. By the time he arrived at school, student-oriented technology had swept college campuses throughout America. Students were using applications like Chegg for help with citations and plagiarism-checking their essays – but few AI writing tools targeted them.
“AI writing is particularly popular with students since they have to turn in a large number of papers per week,” says Boulama. “Sometimes having some initial inspiration helps. An AI writing tool mostly acts as the person to whom you ask, ‘Where should I start?’”
Boulama added a plagiarism checker to Tinq and started talking about his service around campus and online. “Marketing to students is more an art than a science,” he says. “I went to forums like Reddit, posted on TikTok, and also shared with some of my classmates at school. I would say the best strategy, just like in any business, is word-of-mouth.”
But monetizing his product for students was more difficult than for his clients at Otimbi Labs with steady incomes. Students often churned, and when asked in the cancellation forms why they left, they usually said they only wanted to use the product once.
To maintain steady revenue, Boulama created a one-off credit option for Tinq that he still uses today. The credits are slightly more expensive than a day of the monthly subscription and expire after a year. Students liked this pricing model much better, and by mid-2022, 30 percent of Tinq’s customers were either students or academics.
“Entrepreneurs often overlook the student market because students don’t have as much disposable income as other niches,” he says. “Companies that understand the power of marketing to students do it in two ways: One, by providing such insane value that the cost versus value calculation isn’t even necessary. Two, by providing discounts for students and getting them hooked on their products so that when they graduate, they’ll use their products by default.”
In a Class of Its Own
Always looking for new ways to grow Tinq, Boulama whitelabeled his paraphrasing API. This has worked so well that one of the first-ranked websites for the keyword “paraphrasing API” on Google is not Tinq but an application powered by Tinq.
“That’s good for us because they’ll probably want to upgrade their plan,” jokes Boulama.
“Sometimes a technical problem is not something you can monetize. You need to hear people criticize your idea.”
Boulama still manages everything with the help of a coworker based in France who handles customer service while he’s asleep. He believes talking to customers is the most important thing a founder can do and spends much of his free time talking to free and paid customers.
“Before deciding to create a SaaS, I read a lot about what other SaaS founders did,” says Boulama. “The most common advice I saw was to talk to customers. Technical founders like me like to solve every problem we see, but sometimes a technical problem is not something you can monetize. You need to hear people criticize your idea.”
With large startups like Grammarly, Jasper, Quillbot, and GPT-3 competing, the AI writing market has become flooded with heavy hitters. However, Boulama is unphased. He believes few do everything that Tinq does. “Tinq is solving three problems. AI writing and APIs, plagiarism checking, and text generation. Companies that do strictly one thing are not our main competitors.”
Bringing the American Tech Scene Back to Niger
Boulama hopes he can transplant some of the American tech environment back to Niger. The tech ecosystem is growing back in his home country, but he believes it’s too dependent on the government.
“In the US, people understand that if the government can’t fix a problem, an entrepreneur can build a business out of fixing it.”
“I think seeing how the rest of the world makes things happen will lift the Nigerien tech scene. In the US, people understand that if the government can’t fix a problem, an entrepreneur can make a business out of fixing it. In Niger, many create startups hoping to get government funding, and if they don’t get it, they die out because no one will invest.”
But Boulama is still young and his contributions to the world may extend far beyond his SaaS or Niger’s tech scene. He’s always thinking of how he can apply his expertise in AI to the betterment of humanity in space.
“I want to build things at the intersection of rockets and AI,” he says. “For example, we need AI to run single-stage orbit rockets without jettisoning different parts. We also need systems to minimize the effects of space debris on spacecraft. I’m excited to solve these problems.”
Boulama’s success should reassure other founders that even in low-value markets, startups that provide a valuable service still can profit provided they keep overhead low. VC-backed startups with bloated teams and paid acquisitions seldom want to manage the uncertainty of one-off purchases and the low prices demanded by customers like students. Perhaps it is only the bootstrapped startups, unfettered by unrealistic growth goals, that can ever hope to maintain the low prices necessary to sell to those who need them most.
Want to share your bootstrapping story with the world? Enter your details in this form and we’ll be in touch.
