Lead Software Engineer

Epsilon; Bengaluru; On-site; INR 4,000,000 - 6,500,000; Full time

Job summary

Epsilon is seeking a Software Engineer for the Agentic Platform team to contribute to design, delivery, technical strategy, and execution for agent platform systems and to guide end-to-end delivery across loyalty, customer data, messaging, and enterprise AI capabilities.

You will work across Python, Django, TypeScript, React, and AWS to build scalable cloud-native services, agent runtimes, and UI components while mentoring engineers and shaping engineering standards to support growth.

Qualifications

• 8–11 years of software engineering experience with ownership/delivery leadership.
• Proven ability to design and deliver scalable distributed systems across backend, frontend, and runtime layers.
• Strong leadership and mentoring experience for engineers and cross‑service delivery.
• Hands-on experience with Python and Django for production APIs including async workloads (Celery).
• Strong TypeScript/React experience for modern web apps.
• Experience with REST APIs, microservices, containers and cloud‑native architectures.
• Proficiency with PostgreSQL, Redis, and OpenSearch; exposure to document/graph databases is a plus.
• Deep AWS experience (ECS Fargate, RDS, S3, SQS, ALB, IAM, CloudFront, Bedrock).
• Knowledge of asynchronous systems, messaging, and real‑time (SSE/WebSockets).
• Experience with authentication/authorization in multi‑tenant SaaS (Okta/OIDC, JWT).
• DevOps/IaC practices (Terraform/AWS CDK), CI/CD, testing frameworks.
• Exposure to generative AI tech (LLMs, RAG) and agent-based systems preferred.

Responsibilities

• Build and deliver large-scale cloud-native agent platform capabilities on AWS.
• Work across Python, Django, TypeScript, React, FastAPI, Celery, PostgreSQL, Redis, OpenSearch, AWS Bedrock, Strands Agent SDK, MCP, Terraform, and Docker.
• Design, develop, and operate agent runtime services, incl. spec compilation, pre‑flight authorisation, and MCP tool orchestration.
• Build and enhance agent builder and product UIs (Agent Cloud, Core Agent, responsive web) using React 19 shadcn/ui, TanStack Query, and type‑safe OpenAPI clients.
• Implement MCP catalogue, gateway, and federation; integrate domain MCP servers for loyalty, CDP, messaging, and enterprise systems.
• Drive performance, scalability, reliability, security, tenant isolation, governance, and cost efficiency across agent execution and platform services.
• Collaborate with global engineering, product, architecture, AI/ML, and business stakeholders to align technical solutions with business goals.
• Own the end-to-end SDLC: requirements gathering, solution design, development, deployment, observability, and documentation.
• Contribute to observability and quality via OpenTelemetry instrumentation, Grafana/Phoenix tracing, pytest/Vitest/Playwright testing, and CI/CD pipelines.
• Lead/Staff: mentor engineers, lead design reviews, set engineering standards, and drive cross‑team initiatives for runtime performance, MCP federation, orchestration, and multi‑tenant scale.

Skills

Python

Django

TypeScript

React

REST APIs

Microservices

AWS

CI/CD

PostgreSQL

Redis

OpenSearch

Terraform

Docker

Okta/OIDC

JWT

Leadership

Education

Bachelor's or master's in computer science/related field

Tools

Celery

PostgreSQL

Redis

OpenTelemetry

Terraform

AWS CDK

Docker

ECS Fargate

FastAPI

React 19

Job description

As a Software Engineer on the Agentic Platform team, you will contribute to design, delivery, technical strategy, and execution for critically important agent platform systems. Depending on level, you will implement well-scoped features, own end-to-end delivery, lead epics and mentor engineers, or define platform-wide architecture and engineering standards. You will help ensure the Agentic Platform scales securely and reliably to support sustained business growth across Epsilon product lines, including loyalty, customer data, messaging, digital media, and enterprise AI capabilities.

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 stakeholders 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

• Bachelor's or master's degree in computer science, IT, or a related field.
• 8-11 years of experience in software engineering, including 8-10 years in technical ownership and delivery leadership roles.
• Proven ability to design and deliver scalable, distributed systems across backend, frontend, and runtime layers.
• Strong leadership skills with experience mentoring engineers, leading epics, and driving cross‑service execution.
• Hands‑on expertise in Python and Django (DRF/Ninja) for building production‑grade APIs, data models, and async workloads (Celery).
• Strong experience with TypeScript and React ecosystems (React 19 Vite, and monorepos) for modern web application development.
• Solid experience in REST APIs, microservices, containerisation (Docker/ECS Fargate), and cloud‑native architectures.
• Proficiency in PostgreSQL, Redis, and search/indexing systems (OpenSearch); exposure to document/graph databases is a plus.
• Deep experience with AWS services such as ECS Fargate, RDS, S3, SQS, ALB, IAM, CloudFront, and Bedrock.
• Strong understanding of asynchronous systems, messaging, and real‑time communication (SSE/WebSockets).
• Experience with authentication and authorisation (Okta/OIDC, JWT) in multi‑tenant SaaS environments.
• Working knowledge of DevOps and IaC practices (Terraform/AWS CDK), CI/CD pipelines, and testing frameworks.
• Exposure to generative AI technologies (LLMs, RAG, and agent‑based systems) with hands‑on experience being preferred.

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.