Software Engineer, Agentic AI & AWS Systems

Jobtailor; Cape Town; On-site; ZAR 600,000 - 1,000,000; Full time

Job summary

Sanlam is seeking a senior software engineer to own agentic workflows, integrating multi-agent orchestration within AWS and ensure production readiness.

You will work across BFF/microservice delivery, memory/services and incident triage, with Datadog and CloudWatch for observability and incident response. Strong Java/TypeScript/Python skills and a security-first approach are required.

Qualifications

• Relevant Degree or Diploma (BSc Computer Science, IT or equivalent experience)
• 5+ years of experience in Software Engineering
• Production systems experience on AWS or equivalent
• Hands-on experience building with LLM APIs such as Anthropic, Bedrock or OpenAI
• Experience with tool use/function calling, structured output, prompt management and context management
• Demonstrable daily use of agentic coding tools such as Claude Code, Cursor, Copilot agents or similar
• Proficiency in Java, TypeScript and Python
• Ability to read, review and understand code, including agent-written code
• Ability to articulate language differences and build and execution processes
• Understanding of retrieval, embeddings, memory patterns, knowledge graphs and vector stores
• Experience with agent orchestration frameworks or hand-rolled agent loops
• Experience building evaluations and observability for non-deterministic systems
• Awareness of prompt-injection, data-exfiltration and privilege-escalation risks and mitigations
• Ability to reason about data structures, algorithms, scalability, robustness and distributed systems
• AWS experience with Lambda, ECS/containers, Kinesis, DynamoDB, S3, VPC, IAM and Cognito preferred
• Understanding of monitoring, metrics, health, fault tolerance, APIs, authentication, databases, persistence and event-driven architectures
• Security-by-default mindset covering data residency, POPIA, least-privilege RBAC and audit trails

Responsibilities

• Take ownership of new capabilities across the pod's five focus areas and deliver them from concept to production inside Sanlam's AWS tenant
• Build and operate agentic workflows, including multi-agent orchestration, MCP tool integrations, context/memory services and security guardrails
• Integrate agents into engineering flows including BFF/microservice feature delivery, local runners, acceptance tests and incident triage
• Use Datadog, CloudWatch and dead-letter queues for incident triage and resolution
• Author reusable AI skills, prompts and team-context packages for Sanlam engineers
• Measure evaluations, latency, token cost-per-feature and reliability of agent runs
• Feed results into Codeship cockpit metrics
• Participate in code reviews and improve development, testing and operational practices
• Participate in on-call for the platform and handle incidents
• Identify and reduce technical debt

Skills

Java

TypeScript

Python

LLM APIs

Agent orchestration

AWS

Education

BSc Computer Science or equivalent

Relevant Degree or Diploma

Tools

Datadog

CloudWatch

Codeship

AWS Lambda

ECS/Containers

Kinesis

DynamoDB

S3

VPC

IAM

Cognito

Claude Code

Cursor

Copilot

Job description

Sanlam is seeking a senior software engineer to own agentic workflows, integrating multi-agent orchestration within AWS and ensure production readiness.

You will work across BFF/microservice delivery, memory/services and incident triage, with Datadog and CloudWatch for observability and incident response. Strong Java/TypeScript/Python skills and a security-first approach are required.