2 дня назад
Graduate Conversational AI QA (Fintech)
Мэтч & Сопровод
Для мэтча с этой вакансией нужен Plus
Описание вакансии
Текст:
TL;DR
Graduate Conversational AI QA (Fintech): Evaluating AI assistant conversations for accuracy, helpfulness, appropriate handoffs, and regulatory compliance with an accent on Consumer Duty, conduct risk, and customer outcomes. Focus on investigating root causes, designing adversarial test scenarios, and turning evaluation evidence into improvements to models, knowledge content, guardrails, and conversation design.
Location: City of London, United Kingdom; hybrid work with three days in the office
Company
is a FTSE 100 fintech operating across five continents, serving more than 1.4 million customers and handling high-volume financial transactions.
What you will do
- Review client conversations with the AI assistant across chat and digital channels, scoring accuracy, clarity, tone, resolution, handoffs, and compliance.
- Identify conduct risks involving investment advice, product information, fees, margin, personal data, complaints, and vulnerable clients.
- Investigate failed conversations and negative outcomes through root cause analysis across intent recognition, knowledge content, conversation design, model behaviour, and handoffs.
- Design and run test scenarios, including edge cases and adversarial prompts, before and after bot changes.
- Provide evidence-based feedback with example transcripts and recommended improvements to product, knowledge, CX, data, and compliance colleagues.
- Contribute to AI quality measurement, evaluation datasets, scoring criteria, automated evaluation, and wider bot improvement projects.
Requirements
- Degree in any discipline.
- Excellent written English with strong spelling, punctuation, and grammar.
- Genuine interest in AI and how conversational systems behave in real-world situations.
- Strong analytical, problem-solving, customer-focused, and attention-to-detail skills.
- Interest in financial services regulation and the reasons behind Consumer Duty and conduct requirements.
- Previous QA or financial services experience is not required; full training is provided.
Nice to have
- Internship, placement, or part-time experience in customer service, compliance, data, or a regulated environment.
- Hands-on experience with AI or large language model tools.
- A second language for evaluating conversations outside English.
- Exposure to Excel, SQL, Python, research methods, experiment design, or structured quality measurement.
Culture & Benefits
- Structured training in QA, financial services regulation, and AI evaluation with a dedicated mentor.
- Exposure to cross-functional AI projects and opportunities to develop toward QA, AI quality, conversation design, or product roles.
- Hybrid working with three office days per week.
- Private medical cover, life insurance, gym membership contribution, and a flexible benefits package worth 12% of salary.
- Twenty-five days of holiday plus birthday and volunteering days, with the option to buy or sell holiday.
- Enhanced parental leave, onboarding, professional networks, mentoring, volunteering, sports, and social activities.
Hiring process
- Initial training includes CX agent training, product and systems familiarisation, bot evaluation training, and financial services regulation instruction.
- Progression begins with guided evaluations, followed by independent root cause analysis, test scenario development, and presentation of findings.
Будьте осторожны: если работодатель просит войти в их систему, используя iCloud/Google, прислать код/пароль, запустить код/ПО, не делайте этого - это мошенники. Обязательно жмите "Пожаловаться" или пишите в поддержку. Подробнее в гайде →