Staff Software Engineer

Responsibilities:

• Build and deliver large-scale cloud-native agent platform capabilities primarily on AWS, leveraging REST APIs, WebSockets, Server-Sent Events (SSE), microservices, and event-driven architectures to create highly scalable and resilient systems.
• Work hands-on across the technology stack, including Python, Django, TypeScript, React, FastAPI, Celery, PostgreSQL, Redis, OpenSearch, AWS Bedrock, Strands Agent SDK, MCP, Terraform, and Docker, to solve complex engineering challenges and maintain platform excellence.
• Design, develop, and operate agent runtime services, including spec compilation, pre-flight authorisation, skills seeding, MCP tool orchestration, streaming responses, and run package recording.
• Build and enhance agent builder and product UIs (Agent Cloud, Core Agent, and responsive web) and shared frontend libraries using React 19 shadcn/ui, TanStack Query, and type-safe OpenAPI client generation.
• Implement MCP catalogue, gateway, and federation capabilities; integrate domain MCP servers for loyalty, customer CDP, messaging, digital media, and related enterprise systems.
• Drive performance optimisation, scalability, reliability, security, tenant isolation, governance, and cost efficiency across agent execution and platform services.
• Collaborate closely with global engineering, product management, architecture, AI/ML, and business partners to align technical solutions with strategic business objectives.
• Own the end-to-end software development lifecycle, including requirements gathering, solution design, development, deployment, observability, and documentation.
• Contribute to observability and quality through OpenTelemetry instrumentation, Grafana/Phoenix tracing, pytest/Vitest/Playwright testing, and CI/CD pipelines.
• At Lead and Staff levels: mentor engineers, lead technical design reviews, set engineering standards, and drive cross-team initiatives for runtime performance, MCP federation, orchestration, and multi-tenant scale.

Requirements:

• B. E. /B. Tech. /M. Tech. /MCA in Computer Science, IT, or related fields.
• 12+ years as a staff software engineer with expertise in scalable applications, distributed systems, and org-level technical leadership.
• Defines platform architecture (runtime, MCP federation, orchestration, and multi-tenant scale) and drives standards for security, observability, and reliability across teams.
• Expert in at least one area: distributed systems, LLM/agent runtime, or cloud platform engineering.
• Represents engineering in reviews, incidents, and strategic planning.
• Advanced Python and Django (REST Framework/Django Ninja) skills for backend APIs, Celery, and multi-app architectures.
• Proficient in TypeScript and React (React 19 Vite, and pnpm) for modern web apps using component libraries (shadcn/ui, Radix UI, and Tailwind).
• Experienced in designing scalable REST APIs, microservices, containerised workloads (Docker/ECS Fargate), and distributed systems.
• Knowledge of PostgreSQL (pgvector), Redis, and OpenSearch; familiarity with document/graph stores (DocumentDB, Neo4j, and MongoDB) is a plus.
• Extensive AWS experience (ECS Fargate, RDS, S3 SQS, ALB, CloudFront, Secrets Manager, IAM, and Bedrock).
• Skilled in asynchronous processing (Celery, message queues) and SSE/WebSocket streaming for real-time interactions.
• Authentication/authorisation expertise with Okta/OIDC, JWT, and claims-based access in multi-tenant SaaS.
• Hands-on IaC tools (Terraform/AWS CDK).
• CI/CD and DevOps proficiency (GoCD, CodeDeploy, ECR, GitHub/GitLab, automated pipelines).
• Strong testing practices (pytest, Vitest, and Playwright) and adherence to engineering best practices.
• Excellent critical thinking and problem-solving abilities.
• Familiarity with generative AI technologies (LLMs, RAG, MCP, agent orchestration, and agentic AI); hands-on preferred for senior roles.

Nice to have:

• Experience with Strands Agent SDK, AWS Bedrock AgentCore, LiteLLM, or comparable agent runtime frameworks.
• Experience building MCP servers or tool-calling integrations using FastMCP or similar frameworks.
• Experience with agent evaluation (DeepEval, Guardrails AI), LLM observability (Arize Phoenix, Langfuse, OpenTelemetry), and token metering.
• Experience with React Flow (xyflow) for workflow visualisation, CodeMirror for in-browser editors, and declarative UI / SSE event streaming.
• Experience with Expo / React Native for mobile applications.
• Experience with A2A (Agent-to-Agent) protocol, cross-account AWS agent execution, and Step Functions / EventBridge orchestration.
• Working knowledge of Neo4j, Trino, Databricks, or Vertex AI RAG integrations.
• Experience building platforms in privacy-safe, customer data platform (CDP), or marketing technology environments.
• Experience with GoCD, Husky pre-commit hooks, Orval OpenAPI code generation, and monorepo development workflows.

Specialisation Tracks:

• Platform backend Django apps, APIs, Celery, PostgreSQL, tenant isolation, and publishing workflows.
• Agent Runtime / AI Strands SDK, Bedrock, MCP, RAG, evaluation, streaming, and performance optimisation.
• Frontend / UX React apps, shadcn/ui, workflow editors, SSE streaming, type-safe API clients.
• Mobile Expo / React Native companion application.
• Platform / Infra: Terraform, ECS, observability, CI/CD, cross-account AWS.