Staff AI Engineer, Typeform

• Significant experience building and operating machine learning or AI systems in production, with evidence of technical leadership beyond your own projects
• A track record of leading complex engineering initiatives across teams, from ambiguous requirements to measurable production outcomes
• Strong Python and software engineering skills, with the ability to contribute directly to production code
• Experience designing production services and APIs using frameworks such as FastAPI
• Practical experience building generative AI applications using large language models, RAG, tool use, or agentic systems
• A strong understanding of enterprise RAG systems, including retrieval architecture, chunking, embeddings, reranking, evaluation, and monitoring
• Experience defining evaluation approaches and using evidence to guide model, architecture, and release decisions
• Experience with frameworks such as PyTorch, LangChain, LangGraph, or similar technologies
• Strong experience designing and operating cloud systems using AWS, Docker, Kubernetes, Terraform, and continuous integration and deployment practices
• Familiarity with services such as AWS SageMaker or AWS Bedrock
• Experience with event-driven processing, vector databases, and machine learning lifecycle tools such as Kafka and MLflow, or comparable technologies
• Experience establishing observability and diagnosing production issues using tools such as Datadog or OpenSearch
• Sound judgement when balancing delivery speed, quality, reliability, scalability, security, and cost
• The ability to influence technical decisions across teams and build alignment without relying on formal authority
• Experience mentoring engineers and improving the effectiveness of the teams around you

Desirable

• As a Staff AI Engineer at Typeform, you will play a central role in shaping the technical direction of Research Flow and the AI systems that support our broader products
• You will take ownership of complex engineering challenges that span teams, from defining architecture and testing new approaches to delivering and operating production systems. A key part of your role will be connecting individual AI capabilities into a dependable customer experience
• Your scope will span generative AI applications, enterprise RAG systems, agentic workflows, model evaluation, machine learning pipelines, and the infrastructure required to run them reliably at scale
• This is a hands-on individual contributor role with influence beyond a single project. You will lead through technical judgement, delivery, and collaboration, helping teams make sound decisions and building foundations that other engineers can use
• Partner with Product and Engineering leaders to translate Research Flow’s product ambitions into a clear technical direction and delivery priorities
• Lead architectural decisions across AI-assisted study design, adaptive conversations, and research synthesis
• Identify the technical constraints and dependencies that matter most, and help teams address them early
• Define how AI capabilities, data flows, and services work together as the product evolves
• Balance immediate delivery needs with longer-term reliability, scalability, and maintainability
• Take technical ownership of ambiguous problems, from defining the problem and exploring approaches through to production delivery
• Design and build generative AI applications using large language models, RAG, vector search, tool use, and agentic systems
• Stay close to implementation through prototyping, production code, design reviews, and debugging
• Lead initiatives that require coordination across Product, Engineering, Data Science, and Data Engineering
• Build reusable services and APIs that help product teams deliver AI capabilities consistently
• Make pragmatic decisions about when to build, buy, simplify, or stop an approach
• Guide the architecture of machine learning services and workflows using Python, Docker, Kubernetes, and AWS
• Design reliable pipelines for batch and real-time processing using technologies such as Kafka and Airflow
• Establish patterns for retrieval, vector search, model orchestration, and working with structured and unstructured data
• Improve how we manage experiments, model versions, registries, and deployments using tools such as MLflow
• Identify and resolve performance, reliability, and cost bottlenecks across our AI systems
• Help teams choose infrastructure and tools that fit the problem and can be operated sustainably
• Define evaluation strategies and release criteria for generative AI applications, including Research Flow’s conversational and analytical capabilities
• Guide the development of automated benchmarks covering accuracy, relevance, reliability, fairness, latency, and cost
• Establish ways to assess whether AI-generated follow-up questions are useful and whether summaries and insights are grounded in participants’ responses
• Lead improvements to retrieval quality, including chunking, embeddings, context selection, and reranking
• Connect offline evaluation with production monitoring and customer feedback to guide improvements
• Build security, privacy, and safeguards against unexpected model behaviour into system design
• Establish reusable patterns and technical standards for building, evaluating, deploying, and operating AI systems
• Help engineers reason through difficult technical decisions and make trade-offs explicit
• Mentor engineers and support other technical leads in growing their ownership and judgement
• Improve engineering practices across testing, observability, security, incident response, and deployment
• Build alignment around technical decisions through clear proposals, constructive discussion, and evidence
• Evaluate relevant AI research and emerging tools, and help teams adopt what delivers practical value
• Work with Product and Engineering partners to prioritise AI investments based on customer needs, technical feasibility, and business impact
• Help define measurable outcomes for AI initiatives and use them to assess whether an approach is working
• Communicate technical concepts, risks, and trade-offs clearly to technical and nontechnical partners
• Contribute to planning across teams, making dependencies and sequencing clear
• Help shape the broader direction of AI at Typeform through lessons learned from building and scaling Research Flow