Applied Research Engineer

At Similarweb, we build some of the most comprehensive and unique views of how the digital world actually works.

The Data for AI team is a small, specialized group within Similarweb, working closely with a select set of the world’s leading AI organizations (mostly foundational model companies). The team’s mission is to enable these companies to improve their models and AI assistants by applying Similarweb’s data to real-world AI use cases.

The work involves deep collaboration with AI teams and a strong focus on data quality, scale, and applicability to modern machine learning systems. The team operates in a lean, high-ownership environment and plays a direct role in shaping how Similarweb’s data is used in advanced AI products.

Why is this role so important at Similarweb?

Similarweb is a data-focused company, and our unique AI and machine learning capabilities are at the center of our business. In the Data for AI group, we work directly with foundational AI companies that are shaping the next generation of models, helping them leverage real-world digital signals and conversational data to improve performance, coverage, and reliability.

As part of this role, you will design and develop data models that transform petabytes of real-world behavioral and conversational data into high-impact training and evaluation assets. You will research and build new methodologies that allow leading AI organizations to better understand, benchmark, and enhance their models using signals that reflect how people actually interact, search, browse, and engage online.

As an applied research engineer in the Data for AI team, you will operate at the core intersection of large-scale data and frontier AI development. You will turn complex, raw datasets into structured signals, metrics, and model-ready outputs that directly support foundation model training, fine-tuning, and validation. The team’s deeply collaborative and customer-facing nature means you will work closely with strategic AI partners, translating real-world model needs into robust data solutions. Together, you will convert real-world conversations and digital behavior into measurable improvements in model quality and performance.

So, what will you be doing all day?

Your Daily Responsibilities May Include

• Working on high-impact use cases with leading foundational AI companies in a fast-paced production environment
• Applying strategic thinking and strong problem-solving skills, with a clear focus on model performance and business impact
• Transforming large-scale, real-world behavioral and conversational data into high-quality training and evaluation assets
• Understanding partner model challenges and building data solutions that drive measurable improvements
• Leading initiatives end-to-end, from exploration and proof of concept to production deployment at scale
• Owning validation and benchmarking processes for data products and model-impact metrics
• Working hand in hand with senior advisors, including former leaders from NVIDIA and Google, to address cutting-edge challenges in foundation model development
• Supporting the deployment of machine learning and data-driven systems used by top-tier AI organizations
  
  

• Holds a B.Sc. in Computer Science/Mathematics/BioInformatics or any other relevant field - Required (Msc big advantage)
• Has 3+ years of experience with Python
• Strong communicative and verbal abilities to lead and guide customers through the logic of custom-built models
• Has previous experience developing machine learning/ image processing/NLP or similar algorithms
• Familiarity with SQL
• Experience with Big Data tools and cloud infrastructure; PySpark, AWS
• Training or fine-tuning experience using huggingface, pytorch and such - big advantage
• Hands on background in LLMs & AI Agents things like langchain, LlamaIndex, ollama, and such - big advantage