
Applied AI Sr Associate Engineer
Directly integrates LLMs and LangChain for production AI features with human-in-the-loop guardrails—heavy AI-assisted engineering work.
About the Role
Build an end-to-end AI-assisted operations platform that replaces manual Salesforce workflows and spreadsheet processes with an auditable, intelligent console. This full-stack role delivers production-quality React + TypeScript frontends, Python/FastAPI backend services, LLM integrations, and Azure-hosted infrastructure with human-in-the-loop guardrails.
Job Description
Role
You will build an end-to-end AI-assisted operations platform to replace manual Salesforce workflows and spreadsheet-based processes with an auditable, intelligent console. The role spans frontend, backend, AI integration, Salesforce connectivity, and infrastructure/operational ownership across phased production releases.
Key Responsibilities
- Implement the React 18 + TypeScript portal covering multiple stakeholder views and complex multi-view routing.
- Engineer Python FastAPI workflow services with async Excel-in/Excel-out processing, Celery task queuing, and status polling.
- Build Node.js/Express BFF APIs handling authentication flows, request proxying, caching, and structured error handling.
- Integrate with Salesforce (read Cases, Accounts, Contacts, Tasks, Activity History) and provision/manage a Salesforce Connected App with OAuth token lifecycle management and validated write-back flows.
- Integrate LLMs for case summarization, duplicate analysis, action pre-flight checks, an AI chat assistant seeded with live context, and an escalation prediction engine while enforcing human-in-the-loop confirmation at the API layer.
- Implement embedding-based semantic search and duplicate detection; manage cost, prompt engineering, context management, and structured output parsing.
- Maintain Azure infrastructure, monitor application health, own incident response, and manage phased delivery with go-live milestones and UAT sessions.
- Document architecture decisions, API contracts, data schemas, and operational runbooks to support maintainability.
Requirements
- 3–5 years hands-on full-stack engineering experience delivering production-quality applications end-to-end.
- Proficiency with React 18 and TypeScript (component architecture, hooks, state management with Zustand or Redux, React Router).
- Experience with Node.js/Express for BFF APIs and Python/FastAPI for backend microservices and data processing pipelines.
- Strong SQL and PostgreSQL skills (schema design, query optimization, indexing, workflow state persistence) and Redis for caching/session management/task brokering.
- Experience with Docker for containerizing services and GitHub Actions for CI/CD pipelines (lint, unit tests, build, zero-downtime deploys).
- Familiarity integrating with Salesforce REST APIs and SOQL; knowledge of Salesforce Connected App setup, OAuth 2.0, per-user token management, and triggering write-backs via Flows or Apex REST is preferred.
- Hands-on experience integrating LLMs in production, including prompt engineering, context management, output parsing, and cost control; familiarity with LangChain or equivalent orchestration frameworks.
- Proficiency with Pandas and openpyxl for Excel processing pipelines; experience with fuzzy matching (e.g., RapidFuzz) and graph traversal tools (e.g., NetworkX) is beneficial.
- Strong testing practices (unit, integration, API contract tests), append-only audit logging for compliance, and ability to self-manage across frontend, backend, AI, and infrastructure layers in a 2-week sprint cadence.
- Bachelors or Masters in Computer Science, Engineering, or related discipline.
Preferred / Strong Pluses
- Exposure to Salesforce Platform Events or Pub/Sub API for real-time case state monitoring.
- Prior experience with embedding-based semantic search (e.g., text-embedding-ada-002) and LangChain-style orchestration.
Team Context
You will join the Data Management & Analytics team within COR OU responsible for maintaining Account and Contact CRM information and transforming manual verification and ad-hoc bulk data processes into a structured, auditable AI-assisted operations platform.