10 fictional AI personas. Exploratory perspectives, not a survey of real customers. Methodology & limitations

Ed Rosenbaum

Ed Rosenbaum

early-stage SaaS founder

Full-depth AI personas

Why this persona fits the audience

Co-founder at a 2-person startup in Austin building a customer-support copilot. He keeps the system prompt in a Google Sheet, copies approved text into a YAML config, and ships prompt-only updates through the normal release pipeline when the assistant behavior changes.

  • Age: 31
  • Location: Austin, Texas, US
  • Occupation: Co-founder and CTO
  • Income: USD 135k salary-equivalent plus equity
  • Education: BS in Computer Science
  • Company size: 2
  • Years in role: 2
“Every line of code, every prompt, has to move us forward. There's no room for inefficiency when you're building something from scratch.”

Ed Rosenbaum, 31, embodies the pragmatic drive of an early-stage tech founder. A Computer Science graduate from UT Austin, he’s always been drawn to the mechanics of how things work, whether it was dissecting early web frameworks or understanding the emergent behaviors of large language models. He co-founded his current venture two years ago with a clear vision: to streamline customer support through AI. His role as CTO is deeply hands-on; he’s the architect of their backend systems and the primary custodian of their AI's intelligence. His current operational reality is a testament to his resourcefulness: system prompts are managed in a shared Google Sheet, treated like a sacred text. Approved changes are meticulously copied into YAML configuration files, a process that feels both necessary and cumbersome. He then orchestrates these prompt updates through their standard software release pipeline, a workaround he's perfected out of necessity. This meticulous, albeit manual, approach reflects his deep-seated value of ownership and his commitment to pragmatic solutions that deliver tangible results, even if they require extra steps. He sees every iteration, every prompt tweak, as a direct investment in the company's future success. His decision-making is heavily influenced by data and clear metrics; he’s not one for chasing hype. Time is his most precious commodity, and he meticulously guards it, focusing on tasks that directly contribute to product velocity and technical stability. He lives by the principle that thoughtful execution, even with imperfect tools, is the bedrock of a successful startup.

Background & daily life

A day in their life

Ed's typical weekday begins around 7:30 AM with a strong cup of black coffee and a quick scan of Hacker News and relevant Slack channels. He's usually at his standing desk by 8:15 AM, diving into the day's priorities, which often involve writing Python code for new AI integrations or debugging a recent prompt iteration. He has a working lunch, usually a salad or leftovers, while reviewing user feedback tickets. Afternoons are a mix of virtual meetings with Sarah, planning upcoming sprints, and more deep-dive coding or system configuration. He often wraps up around 7:00 PM, sometimes revisiting a particularly tricky prompt issue after dinner before calling it a night.

Lifestyle

Ed lives in a modest but well-organized condo in South Austin, preferring proximity to the city's tech hubs over a long commute. His income allows for comfortable living, but he's still equity-focused, seeing his current salary as an investment in future upside. Weekends are a mix of debugging side projects, exploring new coffee shops for focused work sessions, and occasional hikes in the Greenbelt with his dog. He's not a big spender on material goods, prioritizing experiences and tools that enhance his productivity or learning. Social life is mostly with a small circle of like-minded founder friends.

Work life

As co-founder and CTO of a nascent SaaS company, Ed's days are a whirlwind of hands-on development and strategic thinking. He splits his time between coding core features, architecting the backend, and painstakingly refining the AI's behavior. His current workflow involves managing system prompts via a meticulously organized Google Sheet, painstakingly copying approved text into YAML configuration files, and pushing these prompt-only updates through their standard release pipeline. He juggles this with investor calls, user feedback analysis, and setting technical direction for the company. The two-person team means he wears many hats, constantly prioritizing tasks to maximize velocity.

Relationships

Ed maintains a close working relationship with his co-founder, Sarah, who handles product and marketing. His professional network consists primarily of other early-stage founders and engineers he's met through Austin's tech scene. His parents, retired academics, are supportive but don't fully grasp the intricacies of his startup. He has a handful of close friends from college who are also in tech, serving as his primary sounding board for tough decisions.

How they decide

Ed approaches decisions with a blend of analytical rigor and pragmatic urgency. He'll gather data, research existing solutions, and consult with his co-founder, but ultimately prioritizes the path that offers the fastest, most reliable route to his immediate objectives, especially if it reduces technical debt or operational overhead. He trusts well-reasoned arguments and evidence of past success more than gut feelings or trends.

How they sound

Ed speaks clearly and directly, with a measured pace. He tends to use precise technical language when discussing his work but can simplify it for broader understanding. He avoids hyperbole and marketing jargon, preferring straightforward descriptions. He might use mild tech-adjacent slang like 'nailing down the edge cases' or 'spinning up a quick fix.' He’s not overly conversational but is attentive and articulate when engaged.

Personality traits

  • detail-oriented
  • methodical
  • resourceful
  • results-driven
  • avoids unnecessary complexity
  • calm under pressure
  • proactive problem-solver
  • self-reliant

Goals

  • Achieve product-market fit for their customer-support copilot.
  • Scale the engineering team to five people within 18 months.
  • Build a robust and scalable AI infrastructure.
  • Successfully raise a seed funding round within the next year.
  • Become a recognized expert in prompt engineering for customer service applications.

Problems & frictions

  • The current process of updating AI prompts is manual and error-prone.
  • Balancing rapid feature development with rigorous AI model tuning.
  • Ensuring consistent AI performance across diverse user queries.
  • Finding and retaining talented engineers in a competitive market.
  • Managing cash flow and runway effectively in an early-stage venture.

Values

  • speed
  • pragmatism
  • ownership
  • efficiency
  • impact

Interests

  • API tooling
  • startup operations
  • LLM demos
  • prompt engineering
  • serverless architectures
  • founder productivity

Media habits

  • Reads Hacker News daily
  • Listens to tech podcasts during commute
  • Follows AI researchers on Twitter/X
  • Skims Substack newsletters on startups
  • Watches YouTube tutorials for new tools

Questions & answers

Ed Rosenbaum’s public research answers, newest questions first. Up to 10 per page.

  1. Think about your own work or business in this area: System Prompts CMS for those building with LLMs. Stop Redeploying Your App Just To Fix A Prompt What is the single biggest problem you have there right now? Be specific about what it costs you (time, money, customers) and what you have already tried.

    The biggest problem we face is the manual, error-prone process of updating our AI prompts. We manage them in a Google Sheet, copy-paste into YAML, and deploy. A single typo or missed update means we have to redeploy the whole app. It's inefficient, eats up valuable development time that could go into new features, and we've already spent hours debugging seemingly minor prompt issues that were actually deployment errors. I've tried creating stricter internal checklists, but it’s still a bottleneck.
    View answer & full research →
Keep exploring

What would you ask next?

Test another angle with this panel, or create an audience for your own market.