TixelJobs
N
nanvia muse

Senior Product Development AI Engineer; Bangalore, India

Ann Arbor, MI, USPosted 2w ago
ML EngineerSeniorFull-time

About the Role

We are seeking a Senior Product Development AI Engineer – Full Stack with 5\+ years of experience who can drive end\-to\-end product delivery while operating GenAI, agentic AI and MCP capabilities in production. This role is hands\-on: you will build customer\-facing features across frontend and backend, define non\-functional requirements, and ensure AI integrations are safe, reliable, measurable, and cost\-effective.


**About You – experience, education, skills, and accomplishments**


* Strong web\-focussed development skills
* Effective communicator at all levels, excellent inter\-personal skills, strong business focus
* Highly self\-motivated, confident to work on projects alone as well as in a team
* An enthusiastic approach to extending knowledge and learning new skills
* Strong commitment to quality of work and a good attention to detail



Expert in


* Java 11/17, Spring Boot (REST APIs, microservices)
* Microservices architecture (API contracts, versioning, service decomposition)
* Angular 18\+, NodeJS (micro frontends)
* Scalable system design (event\-driven, distributed systems)
* Non\-functional architecture (latency, security, observability)



Must have experience in


* End\-to\-end product delivery (UI–API–data–production support)
* Software design patterns
* RAG pipelines (chunking, embeddings, retrieval orchestration)
* Prompt engineering (few\-shot, structured outputs, guardrails)
* AI assistants \& agents (tool calling, workflows, memory patterns)
* AI evaluation (Ragas, prompt testing, regression evaluation)
* Production readiness (monitoring, alerting, incident response)
* Debugging distributed systems (logs, tracing, root cause analysis)
* Working with continuous integration environments


**It would be great if you also had**

* AI integration (Azure OpenAI/OpenAI APIs)


* CI/CD pipelines (GitHub Actions, Jenkins, Azure DevOps)


* Vector databases (Azure AI Search, Pinecone, Weaviate, FAISS)


* Responsible AI (prompt injection defense, PII masking, policy enforcement)


* Experimentation (A/B testing, feature flags – Split.io)


* Platform thinking (shared AI services, reusable components)


**What will you be doing in this role?**



Product Development


* Partner with Product, BA, UX and architects to drive solution architecture, technical specifications, and delivery plans.


* Document non\-functional requirements (latency, availability, security, scalability, observability) and ensure features meet agreed targets.


* Build microservices and RESTful APIs using Java 11/17 and Spring Boot; apply clean architecture principles, API versioning, backward compatibility, and secure\-by\-default patterns.


* Build frontend components in Angular 18\+ including state management, modular design, performance optimization, accessibility, and consistent UX patterns.


* Focus on operational readiness (logging, metrics, tracing, alerting).



AI Development


* Implement AI integrations, including RAG patterns, embeddings, prompt orchestration, and prompt safety/guardrails.
* Build AI assistants and agent\-based workflows embedded into product experiences, including tool/function calling, multi\-step reasoning flows, and user feedback loops.
* Write and maintain high\-quality prompts and system instructions; apply prompt engineering best practices, structured outputs, and robust error handling.
* Design retrieval pipelines: chunking strategies, embedding selection, indexing approaches, and relevance tuning; choose and integrate vector databases as appropriate.
* Establish evaluation (AI eval) practices: define success metrics, build offline/online evaluation datasets, implement regression tests, and monitor quality drift in production.
* Ensure responsible AI practices: data privacy, PII handling, prompt injection mitigation, content safety, and secure integration of model endpoints.
* Optimize AI feature performance with respect to latency and cost, batching, streaming responses, fallback strategies, and
Share