Backend Engineer (Generative AI Engineer) - #1160691

Software Engineer (Generative AI Engineer)

About the Role:
We are looking for an experienced Software Engineer with a strong background and a passion for Generative AI to join our growing technology team in our GenAI team.

In this role, you will be instrumental in building a comprehensive Agentic Framework. You will focus on the complex orchestration of LLMs, integrating third-party tools, establishing agent observability, and designing the intelligent backend services that power our AI applications. If you are a builder who thrives on solving complex orchestration problems and loves using AI-assisted tools ("vibe coding") to accelerate delivery, this is the role for you.

Key Responsibilities:

· Agentic Orchestration & Development: Design and build advanced orchestration layers using Python based frameworks (e.g., LangGraph, n8n) and cloud-native AI services (e.g., AWS Bedrock Agents). Develop and onboard new "skills" and third-party API tool calls for autonomous agents.

· Advanced RAG & Cloud Integration: Architect and optimize Retrieval-Augmented Generation (RAG) pipelines. Integrate with LLMs heavily utilizing the AWS ecosystem (Bedrock) and internal government API gateways, managing context windows and embedding strategies.

· Platform & Marketplace Architecture: Design and develop the backend architecture for our AI capabilities marketplace, allowing users to securely publish, share, version, and manage datasets, custom document chatbots, newly developed AI agents and agentic "skills".

· Backend & API Engineering: Architect and develop robust backend services and APIs. Design and manage relational databases (PostgreSQL) to support application state and complex AI workflows.

· Vector Search & Data Handling: Manage and optimize search/retrieval mechanisms using vector databases and search engines (e.g., OpenSearch) to improve data retrieval accuracy and latency.

· Agent Observability & Evaluation: Implement telemetry and observability for AI agents (e.g., tracking token usage, latency, and tool-call success). Build mechanisms to evaluate LLM outputs and mitigate hallucinations.

· AI Guardrails & Enterprise Security: Implement robust security measures around LLM usage, including prompt injection defenses, data masking, and output filtering, ensuring all AI interactions comply with strict government data privacy policies.

· Data Pipelines & Platform Support: Build robust data ingestion and processing pipelines (ETL) to feed our vector databases.

· Full-Stack Collaboration: Work closely with our frontend engineers and UI/UX designers to seamlessly connect complex AI backend processes with our React and Next.js frontends.

Qualifications:

· Experience: 5+ years of software engineering experience, with mandatory and proven proficiency in backend development, API design, and system architecture.

· GenAI/LLM Experience: Hands-on experience building applications using AI middleware/frameworks (LangChain, LlamaIndex, LangGraph) and managed LLM services.

· Cloud Ecosystem: Strong familiarity with cloud service providers. Deep knowledge of the AWS ecosystem (specifically AWS Bedrock, Bedrock Agents, OpenSearch) is highly preferred, though equivalent experience in Azure or GCP is acceptable.

· Database Mastery: Strong experience with Relational Database Management Systems (RDBMS), specifically PostgreSQL, alongside Vector DB concepts.

· Modern Development: A working understanding of modern frontend frameworks (React/Next.js) and ability to leverage AI-assisted coding tools (vibe coding) while maintaining strict code quality, security, and the ability to debug complex integrations manually.

Bonus Points:

· Workflow Automation: Experience with visual node-based workflow builders (n8n, Flowise).

· GenAI Observability: Experience with LLM observability and evaluation tools.

· Government Standards: Experience building applications that comply with strict data privacy and security standards (e.g., IM8 policies).