Publicis Sapient - Manager/Specialist - Java Technology

Company Overview :

Publicis Sapient is a prominent entity operating within the professional services and communications landscape. As part of a global network, the company specializes in delivering integrated marketing, advertising, and digital transformation solutions to a diverse portfolio of multinational clients. By leveraging data-driven insights and cutting-edge technology, we help brands navigate complex digital ecosystems, ensuring seamless consumer engagement and operational excellence across various media platforms.

Role Overview :

The Manager / Specialist - Technology (Java) will serve as a pivotal hands-on technical leader based in Bangalore, responsible for steering the architectural direction and engineering excellence of our digital platforms. You will bridge the gap between high-level business objectives and technical execution, working closely with cross-functional stakeholders, product managers, and engineering teams to deliver scalable, high-performance systems.

This is a player-coach role with ~50% hands-on coding - you will write and review production code, own HLD/LLD, and lead the design of production-grade GenAI/LLM platform capabilities (gateway, guardrails, evaluation, observability) alongside a large-scale client engagement.

Key Responsibilities :

- Architect and oversee the development of complex, distributed systems using Java and Spring Boot, ensuring high availability and performance for enterprise-grade applications.

- Write and review production code (~50% of the time) - Java/Spring Boot, microservices, Kafka - and own HLD/LLD, class design, and NFR trade-offs.

- Lead the design and implementation of microservices architecture to improve modularity, scalability, and deployment efficiency.

- Manage the end-to-end lifecycle of technical projects - alignment with business goals, timelines, quality standards - while coaching, mentoring, and growing high-performing engineering teams.

- Integrate Kafka-based data streaming for real-time processing and seamless communication between disparate system components.

- Drive cloud-native strategies on AWS (EKS, Lambda, S3), optimizing cost, performance, and automation (Docker, Kubernetes, Terraform, CI/CD).

- Design and build production GenAI/LLM platform capabilities : model gateway/routing, fallbacks and failover, semantic caching, and cost/latency-aware model selection.

- Own AI security : guardrails, prompt-injection defense, data-leakage prevention, and safe tool/agent execution boundaries.

- Own GenAI quality engineering : prompt versioning, golden datasets, automated evaluation (RAGAS-style), and release gating for LLM-based features.

- Build agentic workflows : tool-calling / MCP-style integrations, human-in-the-loop and AI-in-the-loop patterns.

- Implement LLM observability : token usage, model failures, latency, and cost monitoring alongside existing Splunk/ELK/Grafana/New Relic stacks.

- Ensure resilience patterns - circuit breaker, rate limiting, retries, distributed caching (Redis) - at microservice and API-gateway level.

- Implement containerization (Docker/Kubernetes) and CI/CD to streamline dev, test, and production environments.

- Translate complex technical concepts into actionable insights for non-technical stakeholders and executive leadership.

Required Skillset :

1. Experience & Architecture :

- 10 - 16 years of relevant experience in software engineering and technical leadership, with a track record of delivering complex digital solutions in a fast-paced, hybrid environment.

- Demonstrated expertise designing and maintaining large-scale, high-traffic, low-latency distributed systems; deep understanding of modern system design patterns, HLD/LLD, SOLID/DRY, and architectural best practices.

- Hands-on solution architect with 2+ large-scale greenfield deliveries.

- Proven ability to lead and mentor diverse engineering teams, fostering technical excellence, continuous learning, and collaborative problem-solving.

- Enterprise security, compliance, and data security at network and application layers.

2. Java, Frameworks & Data :

- Strong proficiency in Java 8+ (17/21 strongly preferred) - concurrency, streams, lambdas, generics, virtual threads.

- Spring Boot / Spring Cloud, microservices, REST, WebClient; reactive programming a plus.

- Kafka / RabbitMQ / SQS, event-driven messaging.

- Redis / Memcache distributed caching.

- SQL + NoSQL modeling, transactions, isolation levels (PostgreSQL/MySQL/Oracle/MongoDB/DynamoDB).

- Python and/or Go a plus.

3. GenAI / AI Platform (must-have) :

- Production experience with LLM integration (OpenAI/Claude/Spring AI), prompt engineering, and streaming responses.

- LLM gateway/routing, fallbacks, failover, circuit breaking, semantic caching, and cost/latency-aware model selection.

- Guardrails, prompt-injection defense, data-leakage prevention, and safe tool/agent execution.

- Prompt versioning, golden datasets, LLM evaluation (RAGAS-style), and release gating.

- Agentic workflows - tool-calling, MCP, human-in-the-loop patterns.

- LLM observability - token usage, model failures, latency, and cost monitoring.

4. Cloud, DevOps & Observability :

- Advanced knowledge of AWS (EKS, Lambda, S3, CloudWatch) and experience deploying containerized apps with Docker/Kubernetes; Terraform and CI/CD (Jenkins/GitLab/GitHub Actions).

- Observability : Splunk, ELK, Grafana, Prometheus, New Relic, CloudWatch; profiling (jProfiler/Yourkit/Visual VM).

- Testing rigor : JUnit 5, Mockito, TDD/BDD (Cucumber), JaCoCo, SonarQube; AI evaluation suites - regression testing of LLM outputs, factual-correctness checks beyond cosine similarity, production quality gates.

Good to Have :

- Local inference / latency-throughput optimization.

- Formal people management (hiring, appraisals, 1 : 1s, career growth).

- AI delivery in regulated environments (BFSI/fintech).

- Experience with MCP servers, agent frameworks (LangChain/LangGraph).

Education :

- Bachelor's / Master's in Computer Engineering, Computer Science, or a related field.

What We're Looking For (Screening Signals) :

- You have personally built and shipped an LLM gateway or agentic AI platform - not just used AI coding tools.

- You can write compilable code live - LLD, class design, edge cases, tests.

- You can explain routing, failover, guardrails, prompt versioning, and evaluation in production terms.

- You've owned resilience patterns (circuit breaker, rate limiting, caching) at scale.

- You've led engineers and can speak to hiring, mentoring, and growth.