About PepsiCo

PepsiCo is one of the world's leading food and beverage companies, with an iconic portfolio that includes beloved brands like Pepsi, Lay's, Gatorade, Tropicana, and Quaker. Operating in over 200 countries, PepsiCo employs more than 300,000 people dedicated to delivering innovative products that bring joy to consumers while advancing sustainability goals. In Europe, PepsiCo is at the forefront of digital transformation, leveraging cutting-edge AI technologies to enhance consumer experiences, optimize supply chains, and drive operational excellence. Our Barcelona hub in Cornella de Ll. is a hub for innovation, where interdisciplinary teams pioneer AI-driven solutions for the food and beverage sector.

Role Overview

As an AI Safety Engineer at PepsiCo in Cornella de Ll., Barcelona, you will spearhead safety initiatives for our LLM- and agent-powered chatbot systems. This senior-level role demands expertise in adversarial testing, risk assessment, and mitigation strategies to ensure AI technologies are robust, reliable, and aligned with business objectives. Reporting to AI leadership, you will collaborate with product owners, engineers, and stakeholders to safeguard high-impact AI applications against misuse, failures, and emerging threats. With broad organizational visibility, this position offers influence in shaping PepsiCo's Responsible AI practices.

Why AI Safety Matters at PepsiCo

In the fast-evolving landscape of generative AI, safety is paramount. PepsiCo's chatbots and autonomous agents power customer interactions, personalized recommendations, and internal efficiencies. Your work will prevent risks like harmful content generation, prompt injections, and unsafe tool executions, ensuring trust and compliance across our global operations.

Key Responsibilities

Your role will encompass technical leadership in AI safety, robustness, and risk management:

• Lead adversarial testing, including jailbreak attempts, prompt injection detection, harmful content generation, system prompt extraction, and agent tool misuse evaluations.
• Conduct comprehensive end-to-end risk assessments for AI-driven chatbots and autonomous systems, identifying hazards, exposure levels, and tailored mitigation strategies.
• Build and maintain AI safety evaluation pipelines with red team test suites, scenario-based testing, and automated stress tests.
• Define and track safety KPIs such as harmful output rates, robustness scores, and model resilience metrics.
• Analyze failure modes like hallucinations, deceptive reasoning, and unsafe executions, designing guardrails to minimize risks.

Additionally, you will develop reproducible experiments for LLM behavior, integrate safety into CI/CD pipelines with MLOps teams, support model governance, and contribute to research on emerging safety methodologies.

Qualifications & Requirements

To succeed, you need a strong technical foundation and safety mindset:

• Master’s degree in Computer Science, Data Science, Machine Learning, or a related quantitative field.
• 4+ years of experience developing or evaluating machine learning systems, particularly LLM- or NLP-based applications.
• Strong knowledge of Generative AI and Transformer-based models; proficiency with PyTorch or TensorFlow, Python, and ML libraries.
• Proven experience in model evaluations, experimentation, or reliability testing.
• Excellent communication skills to translate technical insights into business-relevant recommendations.

Preferred Qualifications

Stand out with experience in adversarial ML, red teaming, LLM agents, risk frameworks like NIST AI RMF or ISO 42001, safety guardrails, reinforcement learning, cloud platforms (AWS, Azure, GCP), and MLOps. Soft skills include problem-solving in fast-paced environments, cross-functional influence, and passion for Responsible AI principles like fairness and transparency.

Benefits & Perks

PepsiCo offers a premium compensation package designed for work-life balance and growth:

• Competitive salary estimated at €65,000–€95,000 annually, based on Food & Beverage sector rates in Spain with a buffer for expertise.
• Performance-based bonuses and stock options.
• Comprehensive health insurance, dental, and wellness programs including gym memberships.
• Generous paid time off, parental leave, and flexible hybrid work in Cornella de Ll.
• Employee discounts on PepsiCo products and perks like free snacks in our modern office.

Career Growth

At PepsiCo, your career trajectory is limitless. This role provides exposure to executive leadership and global projects, with opportunities to advance into AI leadership, product management, or research roles. We invest in certifications, conferences, and internal mobility programs, ensuring you grow alongside our AI ambitions. Mentorship from top data scientists and access to state-of-the-art tools position you for long-term success.

Why Join Us

Join PepsiCo's AI team in Barcelona to make a tangible impact on a global giant while enjoying Spain's vibrant culture. Work on mission-critical AI safety for consumer-facing innovations, collaborate in a diverse, inclusive environment, and contribute to sustainable practices. With our commitment to Responsible AI, you'll champion ethics in tech, backed by a company renowned for innovation and employee development. Cornella de Ll. offers proximity to Barcelona's tech scene, beaches, and excellent quality of life.

Role FAQs

Q: Is this role remote? A: Hybrid model with flexibility, primarily on-site in Cornella de Ll. for collaboration.

Q: What AI tools will I use? A: Latest LLMs, PyTorch/TensorFlow, cloud MLOps, and custom safety frameworks.

Q: Visa sponsorship available? A: Yes, for qualified international candidates.

Q: How does PepsiCo support AI safety research? A: Through dedicated budgets, partnerships, and publication opportunities.

Q: What's the team structure? A: Interdisciplinary with data scientists, engineers, product managers, and ethicists.