Solo founding Julius AI: lessons from six pivots and 2 million users

Executive overview

Most founders treat "no co-founder" as a blocker. Rahul Sonwalkar treated it as a default and built Julius AI — an AI data analyst adopted by all 935 first-year Harvard Business School MBA students — entirely solo before hiring a team.

Solo founding forces speed and directness: no alignment tax on decisions, no co-founder inertia on pivots. But it demands elite convincing skills, generous equity sharing, and deliberate culture-setting from day one.

Conviction and momentum are the solo founder's only leverage — against doubt, against recruits, against investors.

The co-founder question

• Waiting to find a co-founder is one of the most common ways smart founders stall.
• Eight out of ten founders with co-founders are privately working through some misalignment.
• Rahul's early co-founders left within months; he kept building rather than pausing.
• Start making progress first — momentum attracts co-founders, hires, and investors better than a pitch alone.
• The space of exploration is infinite when you're blank-slate; progress narrows it productively.

Hiring as a solo founder

• Hire sooner than a co-founded team would — you can't split functions, so you need multipliers faster.
• Hire people who are better than you at their function; Rahul aimed to be the worst coder on the team.
• The core misconception: solo means people just do what you say. Smart people require convincing, always.
• Find people you can put faith into and walk away from — they should execute and pull others along without waiting for the next instruction.
• Be generous with equity: early team takes meaningful risk, and you have more to give than a three-co-founder company.
• Present two offer options — higher cash vs. higher equity — and treat candidates who ask for even more equity as a green flag.
• Reward people retroactively when they fill shoes beyond their original role.

Culture without co-founders

• When you're three people and solo-founded, the team is ~60% of your cultural foundation.
• Whatever culture exists at four people will be the culture at twenty.
• Hire people you'd spend 14-hour days with willingly — culture becomes an extension of the founder's personality.
• Julius's culture: hacker-first, fluid roadmap, idea to live feature in a day or two.
• Psychological safety is set at the top — Rahul lets the team mock him openly, which makes everything lighter.
• Engineers own product launches publicly: they post on their own Twitter/LinkedIn, appear in launch videos.
• Let the company become part of people's personal brand — it's another form of skin in the game.

Six pivots to Julius: the through line

• Pre-company: a year building mobile apps part-time during COVID; quit only after nothing broke through.
• The thesis from Uber: data is at the core of every business decision; LLMs might let non-engineers query it in plain English.
• Pivot 1: logistics data analysis — nearly impossible enterprise sales cycle; scrapped after three months.
• Pivot 2: Excel Copilot — Microsoft trademark issue forced shutdown, but early usage confirmed demand.
• Pivot 3: Census GPT — huge day-one traffic, zero day-two retention; people don't have real questions about census data.
• Further iterations on public datasets confirmed the technology worked but the data didn't matter to users.
• Key insight: people care about their own data, especially work data — that's what became Julius.
• Learnings compound across pivots even when launches fail; each iteration adds conviction or kills a bad hypothesis.

Convincing as the core skill

• Every solo founder role — fundraising, recruiting, sales — reduces to convincing.
• Deep, almost delusional belief in the thing is the prerequisite; you can't sell what you don't believe.
• Structure your conviction: slim but real chance of working + why we're the right team + why timing is right.
• Reps matter more than technique — the more you convince, the better you get.
• Simplify the message so your team can carry it into recruiting and sales conversations you're not in.
• As a solo founder, you can't be in every interview round; your team must be able to convince candidates too.

Authorship and shared credit

• Equity is table stakes; authorship — genuine shared credit — is what makes people treat the company as their baby.
• Decisions outside your expertise should live with the team; be an unblocker or tiebreaker, not the decider.
• Celebrating team members publicly (launches, posts, videos) ties their personal brand to the company's growth.
• People want to be part of history; give them visible roles in building it.

The bear and bull case for going solo

Bear case:

• The loneliest path in startups — lows hit with no one who truly shares the weight.
• Burnout risk is high when the company needs you on recruiting, marketing, and product simultaneously.
• Without co-founder accountability, the temptation to pivot at the first sign of friction is constant.

Bull case:

• Lowest possible friction to start — no co-founder search, no alignment on space, problem, or solution.
• Critical pivots and trajectory decisions can be made fast, without negotiating with blockers.
• More equity to give early team, which attracts people with genuine skin in the game.
• Startups are a game of outliers; following conventional wisdom averages you down.