DeveloperJobs.io
← Back to all jobs

Job Description

Build connectors and an ingestion pipeline that transform raw knowledge content into a governed, permission-tagged corpus delivered to S3 and Amazon Bedrock Knowledge Bases.

Responsibilities

  • Develop the iKNOW connector for Oracle B2C Service knowledge APIs using a full inventory extract followed by incremental sync based on modified date.
  • Capture article status and answer version during Oracle ingestion.
  • Build SharePoint, OneDrive, and Teams connectors using Microsoft Graph:
    • Enumerate sites and drives.
    • Use delta queries for incremental change.
    • Attach sharing and group membership as permission metadata for every ingested item.
  • Create the Sitecore connector using Sitecore content APIs for public web content in scope.
  • Implement the ingestion pipeline:
    • Land source outputs in S3 using a per-source prefix convention.
    • Perform document conversion and text extraction.
    • Deduplicate documents via content hash.
    • Enrich metadata from taxonomy.
    • Sync ingested content to Amazon Bedrock Knowledge Bases with chunking parameters defined by the architect.
  • Produce inventory extracts for content disposition reviews and counts that feed the end-of-Phase-1 re-baselining.
  • Implement retirement handling so items marked retired or deleted at the source are removed from the index on the next sync.
  • Handle throttling and backoff for Oracle and Microsoft APIs.
  • Ensure idempotent re-runs, plus failure isolation per source so one connector outage does not block the others.
  • Instrument each pipeline stage with CloudWatch metrics and write ingestion runbooks.
  • Pair with client IT during weeks 7–10 to support running, re-running, and extending the pipelines after transition.

Requirements

  • 4+ years building data pipelines in production, including at least 2 years on AWS using Python.
  • Hands-on experience with the Microsoft Graph API for SharePoint and Teams content, including delta queries, permissions, and throttling; or equivalent enterprise content platform experience.
  • Experience ingesting unstructured content (HTML, Office documents, PDF, chat transcripts) into S3 and preparing it for search or vector indexing.
  • Working knowledge of Amazon Bedrock Knowledge Bases data sources and sync behavior, or a comparable vector ingestion path on OpenSearch.
  • Infrastructure-as-code experience with Terraform or AWS CDK, plus CI for deploying pipelines into a client-owned account.
  • Ability to work from a shared backlog in client repositories while producing documentation as work is completed.

Technologies

  • Oracle B2C Service knowledge APIs
  • Microsoft Graph
  • Amazon Bedrock Knowledge Bases
  • S3
  • CloudWatch
  • Python
  • Terraform
  • AWS CDK
  • OpenSearch
  • AWS Glue
  • Step Functions
  • EventBridge
  • Oracle
  • SharePoint
  • OneDrive
  • Teams
  • Sitecore
  • HTML
  • Office documents
  • PDF

Preferred Qualifications

  • AWS Certified Data Engineer – Associate
  • Prior integration with Oracle B2C Service or another Oracle Service Cloud product
  • Prior integration with Sitecore XM or XP content APIs
  • Experience using AWS Glue, Step Functions, or EventBridge for orchestration, and document parsing libraries for Office and PDF formats

Job Location & Experience

  • Location: United States (onsite)
  • Minimum experience: 4 years

When You Join Solvd

  • Shape real-world AI-driven projects across key industries, collaborating with clients from startup innovation to enterprise transformation.
  • Work with a global team with equal opportunities for collaboration across continents and cultures.
  • Thrive in an inclusive environment focused on continuous learning, innovation, and ethical AI standards.

Similar Jobs