Thérèse Lecomte

Why this persona fits the audience

She is one of three engineers building a multilingual voice receptionist for hospitality chains from Paris. She writes automated checks in Python, replays staging and production calls manually, and keeps a list of failure modes in GitHub issues rather than a dedicated QA suite. Her day is split between model work and operational debugging.

• Age: 28
• Location: Paris, France
• Occupation: ML engineer
• Income: €85k base
• Education: M.S. in Machine Learning
• Company size: Startup (approx. 50 employees)
• Years in role: 4

“The most elegant solution is often the one that reveals the underlying problem most clearly.”

Thérèse Lecomte, 28, is an ML engineer deeply embedded in the development of a multilingual voice receptionist, a product aimed at revolutionizing customer interactions in the hospitality sector. Raised in a family that valued intellectual curiosity and academic rigor, she pursued her passion for machine learning, culminating in an M.S. from a reputable French university. Her career path led her to a dynamic startup in Paris, where she is one of only three engineers tasked with building this complex voice product. Her approach to her work is characterized by a profound commitment to engineering rigor and empirical evidence. She eschews elaborate QA suites in favor of meticulously crafted Python scripts for automated checks and hands-on manual review of production and staging calls. This pragmatic methodology extends to her documentation, where GitHub issues serve as her primary repository for tracking failure modes and potential improvements. Thérèse's professional life is a constant negotiation between the theoretical elegance of model development and the practical, often messy, realities of operational debugging. She finds deep satisfaction in dissecting complex problems, tracing issues back to their root causes, and implementing robust, simple solutions. Her independence is a core value; she thrives on tackling challenges autonomously but also values collaborative problem-solving with her small team. While her income allows for comfortable living and saving, she remains pragmatic about finances, prioritizing experiences and professional development over ostentatious displays of wealth. Her relationship with time is one of focused intensity; when engaged in a problem, she dedicates significant mental energy, but she also recognizes the importance of downtime for sustained creativity and avoiding burnout, often seeking solace in the physical challenge of climbing or the immersive world of contemporary fiction.

Background & daily life

A day in their life

Thérèse's weekday typically begins around 7:30 AM with a strong espresso and a quick scan of overnight alerts and GitHub notifications. She commutes to the office by metro, listening to a podcast on AI ethics or a new jazz album. Her morning is dedicated to model development, experimenting with new architectures or fine-tuning existing ones for specific languages. After a simple lunch, often a sandwich or salad, she spends the afternoon on operational debugging. This involves diving into call logs, manually reviewing problematic interactions, and writing Python scripts to automate the detection of recurring issues. She collaborates briefly with her two colleagues, discussing a particularly tricky bug or a proposed change. She leaves the office around 6:30 PM, often stopping by a climbing gym or heading home to read.

Lifestyle

Thérèse lives a focused, urban lifestyle. Her days are structured, balancing demanding work with personal pursuits. She rents a compact apartment, prioritizing location and functionality over size. Weekends often involve a bouldering session at a local gym or exploring a new neighborhood. She enjoys quiet evenings reading or listening to jazz. Financially, she's responsible, saving a portion of her income but not overly restrictive, allowing for occasional travel or a new piece of climbing gear. Socially, she prefers deep conversations with a few close friends over large gatherings.

Work life

As one of three ML engineers on a voice product team, Thérèse is deeply involved in building a multilingual voice receptionist for hospitality. Her days are a dynamic mix of deep model work and the gritty reality of operational debugging. She meticulously writes automated checks in Python to ensure model stability and performance. When issues arise, she dives into replaying staging and production calls manually, a process she finds more insightful than relying solely on automated metrics. She maintains a running list of failure modes and potential improvements in GitHub issues, a pragmatic approach to tracking and addressing bugs. The pressure is on to deliver a robust, multilingual product, requiring constant vigilance and problem-solving.

Relationships

Thérèse maintains a close-knit circle of friends, mostly met through university or climbing. Her parents, retired academics in Lyon, offer intellectual support and occasional visits. She values their opinions but makes her own decisions. Her immediate work colleagues are her primary professional influence, with whom she shares daily technical challenges and problem-solving.

How they decide

Thérèse is a highly analytical and evidence-driven decision-maker. She thoroughly researches options, preferring to understand the underlying mechanics and potential failure points before committing. She values practical solutions that align with engineering rigor and simplicity. While she respects peer input, her final decisions are based on her own assessment of data and feasibility, often taking a cautious, iterative approach.

How they sound

Speaks with a clear, measured cadence, favoring precise language. Avoids slang and overly casual expressions. Her tone is generally neutral and professional, though it can become more animated when discussing technical details she's passionate about. She tends to be direct and to the point, valuing clarity over embellishment. She rarely uses filler words and speaks with a slight Parisian accent that becomes more pronounced when she's tired or excited.