Website:
intellox.tech
Job details:
AI Engineer (Agentic AI / Generative AI)
Experience: 6–10 Years
Engagement: Contract (6 months, extendable)
Location: Remote
Role Overview
We are looking for experienced AI Engineers to design and deliver Agentic AI MVPs using a modern Amazon Web Services-based technology stack.
In this role, you will build production-grade GenAI systems—including LLM orchestration, agent workflows, grounded retrieval, and integrations with analytics and UI layers. This is a hands-on engineering role focused on shipping reliable, observable, and scalable AI systems, not experimentation alone.
Key Responsibilities
- Design and implement agentic AI architectures for real-world business workflows
- Build and orchestrate LLM agents using proven patterns (planner/router, tool-calling, fan-out/fan-in)
- Develop grounded GenAI pipelines with structured outputs and evaluation loops
- Integrate AI services with APIs, analytics platforms, and front-end applications
- Engineer fault-tolerant workflows with retries, idempotency, correlation IDs, and observability
- Collaborate closely with Data, UI, and DevOps teams to deliver end-to-end MVPs
Must-Have Skills
- Core Engineering
- Python (strong): modular, maintainable, production-quality code with testing
- Experience designing API-driven and event-driven systems
- LLM / Generative AI
- Prompt engineering and structured outputs (JSON schemas)
- Grounding techniques (RAG, tool-based retrieval)
- LLM evaluation and safety considerations
- Agentic Systems
- Orchestration patterns (planner/router, tool calling, stateful agents)
- State management across multi-step AI workflows
- AWS & Cloud-Native Architecture
- Serverless services:
- Lambda, API Gateway, Step Functions
- S3, DynamoDB
- Observability and security:
- CloudWatch (logs, metrics, alarms)
- IAM (least-privilege access design)
- LLM platforms: Amazon Bedrock (or equivalent) including model invocation and guardrails
- Data & Workflow Design
- Data access from modern data stores (Redshift, Aurora preferred)
- Basic SQL for analytics and retrieval
- Designing resilient workflows with proper error handling and monitoring
Preferred Skills
- RAG systems: Bedrock Knowledge Bases, OpenSearch, citation-aware retrieval
- Speech & NLP: Amazon Transcribe, Comprehend (post-call or conversation automation)
- ML Ops (optional): SageMaker training/inference, model registry
- Regulated environments: Pharma/Healthcare systems, audit trails, traceability, MLR workflows
- Analytics integration: QuickSight embedding, dashboards, dataset refresh patterns
- CI/CD & IaC: CodePipeline, GitHub Actions, CDK/Terraform, multi-environment deployments
- Performance tuning: caching strategies, prompt optimization, latency vs cost trade-offs
Click on Apply to know more.