Data Scientist - LLM
Role details
Job location
Tech stack
Job description
Role Overview: As a Data Scientist focused on LLM at Gorgias, you will play a crucial role in developing, refining, and enabling our AI Agent, designed to automatically address and resolve shopper requests for our merchant clients. Beyond prompt engineering, you'll be instrumental in scaling the AI Agent's capabilities by developing the right LLM strategies to enable our services. You will work closely with our ML Engineers, Product Managers, and Software Engineers to create, optimize, and implement solutions that enhance the efficiency, accuracy, and functionality of our AI-driven customer support solutions.
What You'll Do
- Design and iterate on LLM-based workflows, including prompting, chaining, and tool usage, to improve the performance of AI agents in production
- Build and improve systems that reliably handle real customer support interactions at scale
- Review large volumes of customer conversations to identify failure modes and areas for improvement
- Develop and maintain evaluation frameworks, including offline testing and online A/B experiments
- Define and track metrics to measure AI agent performance and guide iteration
- Create and maintain datasets for continuous improvement and non-regression
- Collaborate with Product Managers and Engineers to prioritize improvements and ship high-impact features
- Contribute to the design of internal tools and abstractions that enable scalable AI agent development across teams
- Integrate AI agents with internal systems and external e-commerce platforms
- Document methodologies, best practices, and learnings to support team scalability and knowledge sharing
Requirements
- 2-5 years of experience in software engineering, applied AI, or a similar role
- Strong programming skills in Python or TypeScript
- Experience building LLM-powered applications or AI agents in production environments
- Comfortable working with real-world data, including reviewing large volumes of conversations to identify issues and improvement opportunities
- Experience designing and iterating on prompts, including techniques such as chaining, few-shot prompting, or tool usage
- Familiarity with evaluating LLM performance through structured methods, such as offline evaluation and A/B testing
- Product-minded, with a focus on delivering solutions that improve user experience and business outcomes
- Able to balance speed and quality, and iterate quickly based on data and feedback
- Strong communication skills, able to collaborate effectively with engineers, product managers, and other stakeholders
Nice to have
- Experience building internal tools or platforms
- Experience with orchestration or workflow tools, such as Temporal
- Experience working on customer support, SaaS, or e-commerce products
Benefits & conditions
- ️ 5-week vacation plus 2 weeks RTT (We follow each country's appropriate PTO Laws)
- Paid sick leave
- Paid parental leave (16 weeks)
- MacBook Pro
- ️ Personal credit card to buy lunches (we use Swile)
- We provide private health insurance (we use GAN)
- ️ Get up to €700 to set up your workstation at home (working from home should feel breezy)
- Get up to €2000 of learning material and wellness support per year! This includes €1500 for learning material (such as books, courses, and individual coaching sessions) directly linked to your job scope, as well as a €500 wellness budget. Take advantage of these resources to grow in your role and prioritize your personal development and wellness.
- Every quarter, we organize an online company-wide summit to discuss where we're going and strengthen social bonds. Once per year we organize offsite team retreats and company retreats!
AI at Gorgias
At Gorgias, AI is a natural extension of how we work and build. Our teams use it every day to research, write, analyze, code, and craft better customer experiences. Everyone has access to premium AI tools (ChatGPT, Claude, Granola & others) and an annual L&D budget to explore new ones.
The real magic happens when we share what we learn. Our #powerup Slack channel is a digital petri dish of new tools and workflows, and each team has AI champions who showcase fresh ideas during weekly company-wide standups, now practically AI demo sessions.
We see AI not as a replacement for creativity or empathy, but as a multiplier, helping us move faster, think deeper, and serve customers better.