Senior Lead Software Engineer, Full Stack (ai Platform & Knowledge Library) (enterprise Platforms Technology)

Capital One Capital One · Banking · New York, NY +2

Seeking a Senior Lead Software Engineer to build and own an AI platform and knowledge library for enterprise-scale marketing content generation. This role involves designing a RAG-backed corpus, managing embedding search pipelines, leading prompt MLOps, owning AI content compliance APIs, and designing an agentic critic validation stream. The focus is on building shared capabilities and enabling AI systems to generate compliant content.

What you'd actually do

  1. Design and build the Content Moderation Knowledge Library — a business-unit-scoped, RAG-backed corpus covering regulatory requirements, brand rules, claims and disclosures, and marketing context, reusable across channels and product lines
  2. Own the embedding search pipeline — implement and tune retrieval pipelines for grounded AI generation, including chunking strategy, embedding model selection, index refresh, and hybrid search
  3. Lead Prompt MLOps — treat prompts as compiled, versioned artifacts; own the evaluation loop that continuously improves generation quality against labeled production data
  4. Own the AI content compliance and generation platform APIs as shared capabilities across multiple engineering teams — define contracts, manage versioning, and coordinate cross-team integrations
  5. Design the agentic critic validation stream — deterministic parallel validators (sensitive data detection, compliance rules, claims mapping) combined with LLM-as-judge semantic validators (brand voice, coherence, substantiation)

Skills

Required

  • Bachelor's Degree
  • 6 years of experience in software engineering
  • 1 year experience with cloud computing (AWS, Microsoft Azure, Google Cloud)

Nice to have

  • Master's Degree
  • 4+ years of experience in Python
  • 4+ years of experience building and operating retrieval-augmented generation (RAG) systems
  • 3+ years of experience in applied AI/ML or LLM-based systems in production
  • 3+ years of experience with vector search infrastructure (OpenSearch, Pinecone, PGvector, Weaviate, or equivalent)
  • 3+ years of experience designing and owning shared platform APIs
  • 2+ years of experience with AWS services (OpenSearch Service, S3, Lambda, SQS, or equivalent)
  • Experience leveraging interactive AI tooling to accelerate productivity, utilizing capabilities beyond basic code completion

What the JD emphasized

  • AI-powered platforms that operate at enterprise scale
  • applied machine learning, knowledge engineering, and production software systems
  • AI content compliance and generation platform APIs
  • agentic critic validation stream
  • RAG-backed corpus
  • embedding search pipeline
  • Prompt MLOps
  • retrieval-augmented generation (RAG) systems
  • applied AI/ML or LLM-based systems in production
  • vector search infrastructure

Other signals

  • AI-powered platforms at enterprise scale
  • applied machine learning, knowledge engineering, and production software systems
  • AI content compliance and generation platform APIs
  • agentic critic validation stream