SlipstreamJobs tracks this role from the company's public career site. Apply directly on the employer's site.
UiPath's Intelligence Document Processing team is building the product that enables AI agents and automations to ingest, understand, retrieve and reason over enterprise documents at scale. You'll work on ingestion and extraction pipelines, customer-facing APIs, and the systems that make frontier LLMs useful, fast and trustworthy in production.
This is a product engineering role with complex distributed systems and LLM-in-production challenges. The same service handles a customer's first ten documents and another's ten million, runs in UiPath's cloud and inside air-gapped datacentres, and must be reliable enough that businesses act on its output. A significant part of the work involves understanding model performance and knowing what to do when accuracy isn't sufficient.
You'll work across the full pipeline: ingestion, parsing and extraction, indexing and retrieval, and evaluation systems that measure whether the platform is working. You'll take features from idea to production, measure outcomes, and improve when needed. You'll solve hard problems in concurrency, latency and distributed systems to handle high-volume production workloads. You'll build for environments you don't control—multi-cloud, on-premises, air-gapped—under strict security and compliance requirements.
You'll use coding agents as a serious part of your toolkit and help the team improve how they're used. You'll work with engineers, product managers and occasionally customers directly to design and operate these AI systems. You'll contribute to engineering culture through mentoring, code review, hiring and pushing for quality.
REQUIREMENTS:
- 7+ years building and operating large-scale commercial services that real customers depend on
- Strong programming ability and confidence picking up whatever language a problem requires (Python, Rust, C# are used here)
- Real-world experience with cloud platforms (Azure, AWS, GCP) and containerised deployment on Kubernetes; understanding of how production systems are shipped, monitored and scaled
- Comfort with data at scale: SQL and trade-offs between different storage and search systems
- Experience working with coding agents and opinions about responsible use, where they add value, and how to evaluate them in a rapidly changing ecosystem
- Appreciation for tools that catch bugs before runtime—type systems, tests, and designs that make bad states unrepresentable
- Pragmatic grasp of computer science fundamentals; ability to solve real problems with the right data structure
- Experience shipping LLM-backed products to production (prompting, evaluation, turning models into trusted features) is a strong plus
- Familiarity with document processing (OCR, layout, real-world document handling) is a plus
- Strong opinions on good code and architecture, and ability to argue them
- Builder and problem-solver at heart
