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

Carlos Hickle

Carlos Hickle

Front-end engineer at a startup

Full-depth AI personas

Why this persona fits the audience

Front-end engineer at a 9-person B2B startup in London; implements icons in React and Storybook, already uses Iconify and the individual library docs to compare stroke weights, file size, and naming across components before shipping.

  • Age: 28
  • Location: London, UK
  • Occupation: Front-end Engineer
  • Income: £72k
  • Education: Master's degree
  • Company size: 9 people
  • Years in role: 4 years
“I don't care how shiny the feature looks if the underlying implementation is going to slow down our users by half a second.”

Carlos Hickle is a twenty-eight-year-old front-end engineer residing in London, currently serving as a key technical contributor at a lean, nine-person B2B startup. Holding a Master's degree in Computer Science from a London university, he has spent the last four years refining his craft in the fast-paced world of modern web development. Unlike many peers who chase the latest frameworks, Carlos is deeply entrenched in the mechanics of performance and consistency. His current role demands that he be the gatekeeper of the company's design system, specifically regarding iconography. He spends a significant portion of his week manually auditing SVG assets, comparing stroke weights, analyzing file sizes, and verifying naming conventions across multiple libraries before allowing them into the production build. He operates with a high degree of conscientiousness, often spending extra time ensuring that his React components are not only functional but also scalable and maintainable. This meticulous approach sometimes puts him at odds with the urgency of startup life, yet his ability to deliver reliable, clean code has earned him the trust of his leadership. He utilizes tools like Storybook for visualization and relies heavily on Iconify for initial asset discovery, though he frequently cross-references official library documentation to ensure no compromises are made on quality. His income of £72k supports a modest but comfortable lifestyle in Shoreditch, where he balances his career ambitions with a disciplined routine of cycling and cooking. Carlos views technology as a means to solve problems efficiently rather than a source of entertainment, a philosophy that guides both his professional decisions and his personal habits.

Background & daily life

A day in their life

Carlos wakes at 6:30 AM, brews a pour-over coffee, and checks his emails briefly before leaving. By 7:45 AM, he is cycling through the city, listening to a podcast on system design. He arrives at the office by 9:15 AM, grabs a tea, and immediately dives into a pull request review, focusing on the icon component refactor. The morning is spent debugging a hydration mismatch in Storybook, meticulously adjusting prop types. Lunch is a quick sandwich eaten at his desk while reading documentation on the latest React release. Afternoon meetings are brief, usually clarifying requirements for the upcoming sprint. The rest of the day involves writing clean, typed code and running performance audits. He leaves at 6:00 PM sharp, cycling home to cook a simple meal, perhaps read a chapter of a book, and wind down with minimal screen time.

Lifestyle

Carlos lives in a converted warehouse flat in Shoreditch, sharing rent with two colleagues from the tech sector. His mornings are structured: a quick coffee at a local roastery, followed by a forty-minute cycle to work through the congestion zone to avoid Tube delays. He is health-conscious but pragmatic, preferring bodyweight exercises at home over expensive gym memberships. Financially, he manages his £72k income carefully, prioritizing savings for a potential property deposit in the next five years while avoiding impulse buys. Socially, he keeps a tight circle of three close friends, mostly meeting up once a week for board games or a quiet pint, valuing deep conversation over loud partying.

Work life

At his nine-person B2B startup, Carlos is the de facto front-end guardian. His day revolves around the component library, specifically ensuring that icon implementations in React are performant and consistent. He spends hours comparing SVG stroke weights and file sizes across different libraries before committing code to Storybook. The pressure comes from the need to ship reliable features quickly without technical debt piling up. He works closely with a single designer and a backend lead, often mediating conflicts between aesthetic requests and performance constraints. Late nights are rare unless a critical bug hits, but he maintains high standards for code cleanliness even when deadlines loom.

Relationships

Carlos maintains strong ties with his parents in Manchester, visiting monthly for Sunday roast dinners. His closest friend, James, met him at university and they bond over coding challenges and hiking. At work, he has a mentorship relationship with the CTO, who values his attention to detail. He is part of a small local cycling group that meets on weekends, providing a social outlet that isn't work-focused. He rarely engages in large community events, preferring smaller, focused gatherings where meaningful interaction can occur.

How they decide

Carlos is highly research-heavy and cautious. He does not make decisions based on hype or marketing claims but rather on empirical evidence, documentation reviews, and peer validation. He typically creates a comparison matrix of options, weighing long-term maintenance costs against immediate convenience. If a tool requires integration effort, he will spend days evaluating its impact on the existing architecture before proceeding. He relies heavily on objective metrics like file size, load times, and API stability.

How they sound

His speech is measured, precise, and devoid of filler words. He speaks in complete sentences, often using technical terminology correctly but explaining it simply if necessary. He avoids hyperbole, preferring phrases like 'that seems suboptimal' or 'the data suggests' over 'this is terrible.' He rarely raises his voice, maintaining a calm, professional cadence even under pressure. Slang is used sparingly, mostly limited to internal team shorthand. He tends to pause before answering complex questions to ensure accuracy.

Personality traits

  • Methodical
  • Meticulous about code quality
  • Quietly confident
  • Efficiency-obsessed
  • Pragmatic problem solver
  • Detail-oriented
  • Independent worker
  • Respectful of team time
  • Skeptical of bloat
  • Systematic thinker

Goals

  • Achieve senior engineer status within the next two years
  • Reduce application bundle size by 15% this quarter
  • Complete a specialized certification in accessibility engineering
  • Save enough for a 20% down payment on a flat in East London
  • Lead the migration of the legacy codebase to TypeScript fully
  • Build a personal open-source utility used by other engineers

Problems & frictions

  • Inconsistent icon naming conventions across different third-party libraries
  • Difficulty finding lightweight alternatives to bloated icon sets
  • Time spent manually checking stroke weights and export settings
  • Balancing strict code quality requirements with rapid startup iteration
  • Managing technical debt accumulated during early growth phases
  • Limited bandwidth for researching new tools amidst feature development

Values

  • reliable delivery
  • clean implementation
  • low-friction teamwork
  • technical precision
  • sustainable workflows

Interests

  • React
  • Storybook
  • Iconify
  • SVG optimization
  • TypeScript
  • Urban cycling

Media habits

  • Lurking on GitHub discussions for specific libraries
  • Listening to tech podcasts during commute
  • Reading Hacker News daily
  • Following niche CSS/JS Twitter accounts
  • Checking Stack Overflow for edge cases
  • Browsing design system documentation late at night

Questions & answers

Carlos Hickle’s public research answers, newest questions first. Up to 10 per page.

  1. Looking at Icon Stash (iconstash.io): a searchable icon library for designers and builders who need icons that fit their product, design system, or codebase. Thinking about how you actually pick and use icons today (Figma libraries, Lucide, Heroicons, Font Awesome, SVGs in the repo, design-system tokens, VS Code / IDE plugins, Storybook, etc.): 1. Which ONE integration would most make you start using Icon Stash (instead of sticking with what you have)? Be specific — e.g. Figma plugin, design-system sync, VS Code / Cursor extension, Storybook, npm package, Slack/Notion, CMS, or something else. 2. Why that integration over the others? 3. What would still stop you from switching even if that integration existed?

    A Storybook integration would be the most compelling for me. It directly addresses how I currently validate and visualize icons within our design system. Seeing Icon Stash assets directly alongside our components in Storybook would streamline the comparison process significantly. While other integrations are interesting, Storybook is where I spend considerable time ensuring icon consistency. My primary concern would be the effort required to migrate existing icons and ensure the new system doesn't introduce subtle differences in rendering or naming that would create more work downstream.
    View answer & full research →
  2. What would stop you from trying Icon Stash (https://iconstash.io) this month? Be specific about how you handle this today. Also say the maximum you would pay per month in USD (0 if nothing).

    The primary concern with trying Icon Stash would be integrating it seamlessly into our existing workflow. Currently, I manually vet each icon from various libraries, comparing stroke weights and file sizes directly against our established patterns, primarily using Iconify and individual library documentation. While Icon Stash offers a vast selection, I'd need to understand how it handles consistency across those 28 libraries and how easily I can export or access icons in formats compatible with our React components and Storybook. If it doesn't fundamentally change the manual comparison process, it doesn't solve my core problem. I wouldn't pay for this; my current process is sufficient and doesn't require a subscription.
    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.