Drimva

Salesforce

Software Engineering SMTS / LMTS - AI Core (REMOTE)

$172K–$286K

California - San Francisco · Office Tech-Flexible · Full-time · Posted Oct 15, 2026

About the job

Leads development of AI retrieval systems, agent infrastructure, and developer tooling that help production agents find, evaluate, and use enterprise knowledge reliably.

What you'll do

  • Lead retrieval and platform initiatives from design through launch, monitoring, and iteration, including RAG, hybrid search, and agent orchestration.
  • Build answer-quality evaluation, observability, and rollout practices; set and track platform performance, scaling, and reliability standards.
  • Provide technical leadership through architecture decisions, cross-functional delivery, mentoring, code review, and incident response.

What they ask for

  • Production AI or backend systems experience, recently focused on LLM-powered applications; expertise in RAG and hybrid-search retrieval.
  • Expertise building developer tooling for production agents and owning platform initiatives through design and adoption.
  • Demonstrated technical leadership and delivery of complex projects with stakeholders, dependencies, or phases.
  • Strong backend engineering skills in a listed language, plus SQL and RDBMS experience; hands-on AI application lifecycle experience.

Nice to have

  • pgvector or other vector indexes; content ingestion, transcript processing, evaluation suites, or voice/real-time agents.
  • Enterprise SaaS, B2B marketing or sales technology, Salesforce/CRM integrations, or startup experience.

Pay

$172,500 - $260,100 annually (base salary); $207,800 - $285,800 annually (base salary)

Tools

RAG, Ruby on Rails, Python, JavaScript, Go, SQL, PostgreSQL, pgvector

AI focus

Builds LLM-powered retrieval and agent infrastructure, including RAG and hybrid search, production-agent tooling, answer-quality evaluation, observability, and safe model and prompt changes. Uses coding agents for implementation, testing, refactoring, and documentation.

Model-written summary · the original posting is authoritative

First seen Oct 16, 2026 · Last verified Nov 6, 2026 (UTC)