logo[мetahunt]
> Djinni

Software Architect

SouthRivers Data
format:Remote
aws bedrockai agentspythonaws lambdaaws s3api gatewayaws ecsaws fargateaws iamcloudwatchterraformragvector searchlanggraphcrewaillamaindex
mcpopensearch
englishB2
experience6+ years
locationToronto, Canada
> full description

SouthRivers Data is a data & AI engineering partner. We strengthen the teams of consultancies and end clients across the EU, UK, and North America with senior data and AI engineers.

 

We're looking for an AWS AI Architect with strong Agentic AI experience to join an AI PoC/MVP project for a large Canadian insurance company.

 

About the project:

  • Design and build an AI PoC/MVP on AWS
  • Focus on AI agents / agentic workflows and enterprise GenAI architecture
  • Greenfield project with significant ownership over solution architecture and technology decisions
  • Direct work with the client's English-speaking technical and business teams
  • Full-time, long-term potential
  • Location: Toronto / Canada — primarily remote, with a possibility of spending a few weeks onsite at the client's office
  • Potential start date: September 15

     

Requirements:

  • 6+ years in software, data, ML, or cloud engineering, with strong solution architecture experience
  • Strong hands-on experience designing and building GenAI / Agentic AI solutions
  • Solid AWS architecture experience
  • Amazon Bedrock and related GenAI services
  • Experience with AI agents: tool calling, orchestration, memory/state, multi-step workflows, and agent-to-system integrations
  • Experience with at least one agent framework such as Strands Agents, LangGraph, CrewAI, LlamaIndex, or similar
  • RAG architectures, embeddings, vector search, and knowledge retrieval
  • Integration of LLMs/agents with enterprise APIs, databases, and internal systems
  • Python and experience building production-grade AI services
  • AWS services such as Lambda, S3, API Gateway, ECS/Fargate, IAM, CloudWatch
  • Understanding of security, identity, access control, data privacy, and guardrails for enterprise AI systems
  • Infrastructure as Code: Terraform or CloudFormation
  • English B2+ — daily communication with the client, no interpreter

 

Nice to have:

  • Amazon Bedrock AgentCore experience
  • MCP / agent tool integration experience
  • Multi-agent architectures
  • Agent observability, evaluation, tracing, and production monitoring
  • Amazon OpenSearch / vector databases
  • Experience taking GenAI solutions from PoC to production
  • Experience in insurance, financial services, or other regulated industries
  • AWS certifications (Solutions Architect, Machine Learning, or similar)
  • Canadian visa

 

Interview process:

  1. Recruiter interview
  2. Technical interview
  3. Possible additional interview with an AWS / AI engineer