Principal + Staff Software Engineers
- Compensation
- Not disclosed
- Location
- Remote
- Work setup
- Remote
- Seniority
- Principal
- Employment
- Not specified
- Experience
- 10+ years
- Category
- Not specified
- Posted
Skills
Must have
- Have 10+ years of software engineering experience with a focus on distributed systems, infrastructure, or data architecture.
- Are a Python expert—capable of building frameworks and performance-critical services from scratch.
- Have deep Kubernetes expertise; you can design, deploy, and debug at scale and could teach others how to productionize it securely.
- Are fluent in Postgres—you understand query planning, partitioning, and tuning for high-throughput environments.
- Are obsessed with clean, scalable architecture and can lead design reviews that shape how entire systems evolve.
- Have experience in high-performance data or AI/ML systems—especially those involving retrieval pipelines, embeddings, or hybrid workloads.
- Thrive in fast-moving, ambiguous environments where technical depth and judgment matter more than process.
Nice to have
- Experience building or scaling LLM-powered or RAG systems in production.
- Familiarity with open-source orchestration frameworks, vector databases, or hybrid cloud infrastructure.
- Contributions to open-source projects in Python, Kubernetes, or distributed systems.
Full description
Unstructured is defining the standard for enterprise data transformation in the age of LLMs and generative AI. In just two years, we've raised over $65M from world-class investors, including Menlo Ventures, Bain Capital, Databricks, NVIDIA, Microsoft, and IBM.
Our open-source toolkit has been downloaded 97M+ times and is used by 90% of the Fortune 1000. We power production AI workflows across commercial and federal sectors — transforming PDFs, HTML, Word docs, images, emails, and more into AI-ready data pipelines that scale.
We're not just building tools; we're building the backbone of generative AI and the infrastructure that unlocks intelligence across industries.
About the Role
Unstructured is looking for Staff and Principal Software Engineers to define the architectural foundation for how enterprises process and transform unstructured data for large-language-model (LLM) applications.
This is a high-impact, hands-on role for someone who thrives in deep systems work—someone who could write the book on Kubernetes, lives and breathes Python, and understands Postgres at the index and transaction-log level. You’ll own the technical direction for our core platform, guiding how data moves through every layer—from ingestion to transformation to retrieval.
You’ll partner closely with the Head of Engineering, Head of Infrastructure, and product leaders to ensure Unstructured’s systems are performant, resilient, and ready to support the next generation of AI-driven workloads.
Location(s) Preferred: San Francisco, Bay Area, San Jose, Seattle, Sacramento
##
What You’ll Own & Drive
- Define and evolve the end-to-end architecture for Unstructured’s data transformation and retrieval platform.
- Build and scale distributed systems that process massive volumes of unstructured data across diverse formats and sources.
- Serve as the company-wide authority on Kubernetes orchestration, cluster design, performance tuning, and reliability.
- Lead Python architecture and best practices—ensuring performance, modularity, and maintainability across services.
- Design and optimize Postgres schemas, queries, and indexing strategies to support large-scale metadata and retrieval pipelines.
- Mentor senior engineers through design reviews and code guidance, raising the bar for technical excellence across the org.
- Partner with the infrastructure and product teams to translate research prototypes into production-grade systems.
- Evaluate emerging technologies and open-source tools in LLM infrastructure, retrieval, and orchestration—deciding where and how to integrate them.