She left a Harvard PhD to fix the most dreaded HR form in America
The year was 2018. Deborah Hanus was deep inside a Harvard PhD program in computer science and machine learning - writing code, studying reinforcement learning, collecting fellowships. She had three degrees from MIT already. An NSF grant. A Fulbright. A software engineering internship at Google. On paper, she was exactly where you'd expect someone like her to be.
Then six of her close friends tried to take leave from work.
What she watched unfold was not a paperwork problem. It was a system-wide failure. Highly educated people with good jobs at good companies - people who could navigate immigration forms, mortgage applications, and tax audits - could not figure out how to take parental or medical leave without losing their minds or their paychecks. The compliance rules varied by state. The forms were incomprehensible. The HR departments were overwhelmed. The employees were on their own.
Hanus ran the Jeff Bezos "regret minimization" test on herself and published the results on her personal blog: "How Regret Minimization Changed My Life." She moved to San Francisco. She founded Sparrow. She did it alone.
Leave is complicated and stressful. It touches so many aspects of the company - legal compliance, insurance, state agencies, payroll, HRBPs, managers and employees. Everything is always changing, and no one has the data they need when they need it.
- Deborah Hanus, CEO & Co-Founder, SparrowWhat makes this founding story unusual is the company she built. Sparrow is not a self-service app. It is not a compliance checklist generator. It is a high-tech, high-touch leave management operation: AI automation layered under actual human Leave Specialists who handle every case from start to finish. Parental leave. Medical leave. Family leave. The alphabet soup of FMLA, CFRA, SDI, PFL. The 50-state compliance matrix that changes every time a legislature convenes.
The pitch to employers is blunt: you're currently spending 20-40 hours of HR time on every single leave. Sparrow handles it. One customer saved $2.5 million in the first year on a $250,000 investment - a 10x return. Across the platform, Sparrow has saved its clients over $350 million in payroll costs. The formula works because Hanus understood from her research background that compliance isn't just a rule set. It's a data problem.
The regret minimization framework
Before leaving Harvard, Hanus asked herself the Bezos question: "When I'm 80 years old, will I regret not trying this?" She wrote about the decision publicly. The blog post became a kind of manifesto for her generation of founders - the ones who left graduate programs to build companies, not because it was safe, but because not trying felt riskier.
The solo founder bet
Raising venture capital is hard. Raising it as a solo technical founder with no prior startup experience, in a category nobody was romanticizing (HR compliance software), is harder. Hanus did it twice. The Series A - $20 million - came from WndrCo, the fund co-founded by Jeffrey Katzenberg, the man who built DreamWorks. She raised it solo, before hiring a C-suite, before building a proper executive team. The Series B - $35 million, announced July 2025 - came from Silver Lake Waterman, with WndrCo re-upping.
Between the two rounds, Sparrow's revenue grew 14x. The team scaled 3x in a single year, reaching 130+ employees spread across 30+ states. The customer roster became a roll call of the tech industry: OpenAI, Reddit, Bumble, Headspace, Oura, Chime, Gong, Handshake, Eightfold.AI. Companies that pride themselves on moving fast hired Sparrow specifically because leave is the one process you cannot rush, cannot automate away entirely, and cannot afford to get wrong.
I want to create a culture and place where all of the incentives are in the right place for people to really enjoy coming to work and be excited about what they're doing.
- Deborah HanusFrom reinforcement learning to reinforcing HR policy
The jump from academic AI to HR tech is stranger than it sounds, and Hanus rarely lets people forget her technical roots. At PyCon Colombia 2018 - the same year she was deciding whether to finish her PhD - she gave a keynote on machine learning. At QCon New York the year before, she was talking to engineers about systems design. She maintained a GitHub profile (dhanus) with repositories including one built to predict Oscar winners from scraped datasets.
The machine learning background did not sit dormant. Sparrow's platform uses AI to automate the procedural grunt work of leave management: form routing, deadline tracking, state agency communication, payroll calculation. The human specialists handle what AI still cannot: the employee who is frightened, the manager who doesn't know what questions they're allowed to ask, the edge cases that no compliance engine has been trained to recognize.
Before San Francisco, there was Cambodia. Hanus held a Fulbright Student Fellowship, conducting research on education and employment. The experience shows up in how she talks about work: less as an economic transaction, more as a site of meaning and care. Sparrow's stated mission - to support employees through life's critical moments - is not marketing language for her. It's the reason the company exists.
Sparrow: Leave, handled.
Sparrow manages the full lifecycle of employee leave - from the first request to the final return-to-work - combining AI automation with dedicated human Leave Specialists. It operates across all 50 US states and Canada, navigating a compliance landscape that changes every legislative session.
Building culture as a solo act
In 2023, Hanus was named in Transform 10: Most Inspiring Leaders of the Year, recognized specifically for scaling Sparrow from 50 to 130+ employees while maintaining the company's culture. The citation is more interesting than the award. It's hard to build culture when you're the only founder, when your team is spread across 30 states, when your company's core function is navigating the most stressful moments in employees' lives. Hanus has been explicit about her approach: hire for mission alignment first, and make sure the incentives point in the right direction.
She speaks at SHRM conferences - the professional heartland of HR practitioners. At SHRM25, she was on a panel called "Shaping Tomorrow's HR: A Thought Leadership Panel on Emerging HR Tech Solutions." This is not the TED Talk circuit. These are the people who actually process the forms. Getting in front of them, understanding what they're dealing with daily, is how Sparrow's product keeps improving.
Awards & Honors
- Transform 10: Most Inspiring Leaders of the Year (2023) - Recognized for scaling Sparrow's culture through 3x team growth
- 23 Inspiring Women to Watch in 2023 - TechSee
- Fulbright Student Fellowship - Research in Cambodia on education and employment
- NSF Graduate Research Fellowship - National Science Foundation grant for PhD research
- Intel/ACM SIGHPC Computational Data Science Fellowship - Awarded during Harvard PhD
- Great Place To Work Certified - Sparrow certified under her leadership
Timeline
Deborah Hanus on Video
In the News
The details that don't fit elsewhere
Holds three degrees from MIT (two BS, one MEng) and a PhD from Harvard - all before founding her first company.
Conducted Fulbright fellowship research in Cambodia on education and employment before pivoting to tech.
Built a machine learning model to predict Oscar winners by scraping and analyzing film datasets - the project is on her GitHub (dhanus).
"How Regret Minimization Changed My Life" - her personal blog post documenting the decision to leave Harvard and start Sparrow became a founding-story touchstone.
One Sparrow customer saved $2.5 million in their first year on a $250,000 investment - a 10x ROI on leave management software.
Sparrow's customer list includes OpenAI, Reddit, Bumble, and Gong - the company that makes the software Sparrow itself uses for sales calls.