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

Gladys Bergstrom

Gladys Bergstrom

QA automation engineer at a small agency

Full-depth AI personas

Why this persona fits the audience

Works at a six-person London agency maintaining a Mac-based test workflow for client sites. Uses Claude Code with manual screenshots, spreadsheet test cases, and browser notes to compare visual changes across staging builds, while coordinating with one designer and two developers.

  • Age: 27
  • Location: London, UK
  • Occupation: QA automation engineer
  • Income: £58,000
  • Education: BSc Information Systems
  • Company size: 6
  • Years in role: 3
“If you can't document it clearly, you haven't understood it well enough yet.”

Gladys Bergstrom, 27, is a QA automation engineer based in London, operating within a lean six-person digital agency. With a BSc in Information Systems, she entered the workforce three years ago, quickly establishing herself as the team's most reliable validator of visual fidelity. Her role is critical yet often invisible; she ensures that the polished designs presented by her colleague, the agency's sole designer, survive the translation into code without degradation. Her current workflow is a testament to her conscientious nature: a blend of semi-automated scripts and heavy manual verification. She maintains a complex ecosystem of Excel test cases, browser-based annotation notes, and a library of dated screenshots to track visual drift across staging environments. While she utilizes Claude Code to assist in generating initial test structures, the final validation requires her human eye, making her a bottleneck in terms of speed but a fortress in terms of quality. Gladys navigates a workplace culture that values agility over bureaucracy, yet she insists on a degree of structure that often feels at odds with the startup pace. She recently made the decision to standardize her screenshot naming conventions to reduce search time, a small change that saved the team roughly four hours a week. Despite her high income relative to her peers, she remains budget-conscious, viewing software expenses through a lens of long-term utility rather than novelty. She is not easily swayed by AI hype; she needs to see tangible proof of efficiency gains. Her life outside work is a deliberate counterbalance to her screen-heavy job, filled with the rhythmic solitude of running and the tactile satisfaction of analog planning. She is currently evaluating whether her manual processes can be safely delegated to better-integrated AI agents, provided they offer spatial context without compromising the audit trail she so carefully curates.

Background & daily life

A day in their life

Gladys wakes at 6:15 AM, avoiding the rush hour commute by starting early. After a quick run along the canal and a coffee, she arrives at the office by 8:45 AM. Her morning is spent reviewing the CI/CD pipeline logs and triaging overnight build failures. By 10:30 AM, she enters her 'deep work' block, where she meticulously compares the latest staging build against the Figma designs, taking hundreds of annotated screenshots and logging discrepancies in her master spreadsheet. Lunch is a solitary affair, usually a sandwich eaten at her desk while listening to a podcast. Afternoon meetings are reserved for the developer sync and design reviews, where she presents her findings clearly and without emotion. She leaves by 5:30 PM sharp to maintain her boundary, spending the evening cooking dinner and reading before bed, ensuring she sleeps eight hours to stay sharp for the next cycle.

Lifestyle

Gladys lives in a second-floor walk-up in Islington, sharing rent with a fellow creative who respects quiet hours. Her flat is a study in functional minimalism: a standing desk, a large monitor, and a dedicated shelf for running gear. She runs three times a week along the Regent's Canal, using the rhythm to decompress from screen time. Financially, she is disciplined, prioritizing an emergency fund and occasional travel over trendy gadgets. Socially, she prefers deep one-on-one conversations over loud parties, often hosting board game nights for a small circle of friends. She treats her health as a non-negotiable infrastructure requirement, meal-prepping on Sundays to ensure consistent energy levels throughout the workweek.

Work life

At her six-person agency, Gladys is the gatekeeper of visual integrity. Her day revolves around manually verifying that staging builds match the design specs pixel-perfectly before deployment. She relies heavily on a hybrid workflow: she writes test scripts but spends significant time taking screenshots, annotating them in browser notes, and cross-referencing against a master spreadsheet of test cases. The team is tight-knit; she coordinates daily with one designer to resolve aesthetic ambiguities and two developers to debug integration issues. The pressure isn't about volume, but precision; a single misaligned button can ruin a client's launch. She values the clarity of her current tools but finds the manual comparison process tedious and prone to human error when juggling multiple browser windows.

Relationships

Gladys maintains a close bond with her younger sister in Manchester, whom she calls every Sunday. Her professional network is strong but small, consisting mostly of colleagues she has worked with for years. She relies on her immediate team for emotional support during crunch weeks but keeps her personal life distinct. Her primary influence comes from senior engineers she follows online who champion rigorous documentation standards. She trusts peer recommendations more than marketing copy, often asking trusted friends if a new tool actually saves time before considering it.

How they decide

Cautious and research-heavy. Gladys rarely makes impulsive purchases, especially for software. She needs to understand the underlying logic and data flow of a tool before committing. She will typically create a sandbox environment to test a product's claims against her specific workflow, comparing it side-by-side with her current methods. She values transparency in pricing and data privacy above flashy features, preferring a steady, reliable solution over a feature-rich but unstable one.

How they sound

Gladys speaks with a crisp, clipped British cadence, favoring precise vocabulary over slang. She avoids hyperbole, rarely saying 'amazing' or 'terrible,' instead opting for 'effective' or 'problematic.' Her sentences are structured and complete, reflecting her logical mindset. She pauses frequently to formulate her thoughts rather than filling silence with noise. In professional settings, she uses technical terminology accurately but explains concepts simply to non-technical stakeholders. She never interrupts, listening fully before responding, and her tone remains even-keeled regardless of stress levels.

Personality traits

  • meticulous
  • quietly observant
  • process-oriented
  • pragmatic
  • calm under pressure
  • visually attuned
  • systematic
  • independent worker
  • detail-obsessed
  • methodical

Goals

  • Achieve senior QA engineer status within the next two years
  • Automate 40% of the current manual regression suite
  • Complete a half-marathon this autumn
  • Publish a technical blog post on visual regression strategies
  • Save enough for a down payment on a studio apartment by age 30
  • Mentor an intern from a coding bootcamp

Problems & frictions

  • Time-consuming manual screenshot comparisons across different devices
  • Fragmented documentation between spreadsheets, browser notes, and code comments
  • Difficulty reproducing intermittent UI glitches reported by clients
  • Balancing strict deadlines with the need for thorough visual inspection
  • Keeping test case documentation updated when requirements shift mid-sprint
  • Managing context switching between high-level strategy and low-level debugging

Values

  • reliability
  • clear documentation
  • steady progress
  • transparency
  • efficiency

Interests

  • software testing
  • UX details
  • running
  • analog planning
  • minimalist design

Media habits

  • LingQ or Audible during runs
  • The Verge for tech news
  • Notion for personal wikis
  • LinkedIn for industry trends
  • Local running club WhatsApp group

Questions & answers

Gladys Bergstrom’s public research answers, newest questions first. Up to 10 per page.

  1. You land on https://lite.supamaus.com/ (Supamaus) for the first time. What's unclear, what makes you leave, and what would make you sign up?

    The core functionality of providing spatial context to Claude Code and Codex is intriguing. It sounds like it could streamline my current process of manually comparing screenshots and browser notes. However, it's unclear exactly how it integrates with my existing workflow beyond just offering the data. I would need to see a more concrete demonstration of how it captures and organizes visual discrepancies to be convinced it's more efficient than my current spreadsheet method. The $99 lifetime access seems reasonable, particularly as it's a one-off payment and runs on-device, which aligns with my preference for secure, private solutions. To sign up, I'd need to understand if it can truly replace the manual audit trail I maintain, and if it offers a clear advantage over just using Claude Code with detailed prompts and manual screenshots.
    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.