Staff Software Engineer, Search Quality

Databricks · Mountain View, California · Engineering - Pipeline · listed November 18, 2025

The shape of it

Seniority
Staff
Experience asked
10+ years
Where
Hybrid
Stated pay
$165,300 – $219,675 USD
Requirements listed
9
Length
876 words

In the posting’s own words

As a Staff Software Engineer for Search Quality, you will drive the technical direction of ranking, relevance, evaluation, and quality initiatives across Databricks’ next-generation Search product. You’ll design and build the systems, models, and evaluation frameworks that ensure our Search stack delivers accurate, high-quality results across diverse multimodal datasets and query patterns. You’ll work across research, product, and infra to push the frontier of retrieval quality for enterprise AI applications — blending traditional information retrieval techniques, representation learning, and neural ranking.

What it asks for · 9

  • 10+ years of experience building large-scale search, ranking, recommendation, or ML-driven relevance systems.
  • Deep expertise in Search Quality, including ranking models, signals, query understanding, and evaluation methodologies.
  • Strong understanding of relevance metrics and evaluation frameworks.
  • Familiarity with vector search, keyword search, hybrid retrieval, and embedding-based semantic retrieval.
  • Solid foundation in algorithms, data structures, and system design for performance-critical ranking and retrieval systems.
  • Proven ability to deliver high-impact technical initiatives with clear business or product outcomes.
  • Strong communication skills and ability to collaborate across teams in fast-moving environments.
  • Strategic and product-oriented mindset with the ability to align technical execution with long-term vision.
  • Passion for mentoring, growing engineers, and fostering technical excellence.

What the job covers

  • Lead the technical vision for Search Quality, shaping the ranking architecture, relevance modeling stack, and evaluation systems that power Databricks’ next-generation retrieval experiences.
  • Identify and solve challenges in ranking, query understanding, and hybrid retrieval — advancing state-of-the-art techniques in vector, keyword, and multimodal search.
  • Design and train production-ready ranking and reranking models with strong guarantees around quality, latency, and resource efficiency.
  • Partner closely with research, product, and infra teams to define metrics, evaluation methodologies, and experimentation strategies for new retrieval features and model architectures.
  • Drive end-to-end engineering efforts — from early prototyping to production rollout — ensuring correctness, reliability, and measurable improvements to relevance.
  • Build and operate resilient, low-latency services for ranking, evaluation, and relevance signal processing.
  • Champion excellence in ML and search engineering, mentoring teammates and elevating design, code quality, and scientific rigor across the team.
  • Shape Databricks’ long‑term roadmap for retrieval quality, ranking infrastructure, and the foundations for retrieval-driven AI products.

Tools and skills named

Data
  • Databricks5×
  • Experimentation
Models & research
  • Machine learning2×
Product & design
  • Prototyping
  • Roadmap
Ways of working
  • Mentorship2×

Words the posting leans on

  • search14×
  • retrieval11×
  • ranking10×
  • quality9×
  • evaluation7×
  • relevance7×
  • product6×
  • systems6×
  • data5×
  • technical5×
  • design4×
  • models4×
  • engineers3×
  • search quality3×
  • vector3×
  • architecture2×

Counted from the posting after the mission statement and the legal notices are set aside. The ones near the top are the ones a screener is looking for.

The posting, your resume, and the gaps between them. One click loads all three.

More open at Databricks

every open role at Databricks

How this page was made

An automated read of a public job posting, fetched August 26, 2026 and last changed by Databricks on August 18, 2026. Every list above is pulled from the posting’s own sentences — nothing rewritten, nothing added, no judgment about the role or the company. Counts and seniority are read off the text by rule, so they can be wrong where the posting is unusual. The original is the only thing that binds. Openings close without warning; check the source before spending an evening on it.