About the Role :

We are building AI, Agents, and Agentic AI capabilities to drive measurable business outcomes across functions such as customer acquisition, servicing, collections, risk, operations, productivity, compliance, and internal knowledge workflows.

We are looking for an AI Engineer to develop and implement enterprise-grade AI applications, multi-agent workflows, and GenAI platforms. This role is ideal for a strong hands-on engineer who combines software engineering discipline, LLM / GenAI expertise, agent orchestration knowledge, and production deployment experience.

The person in this role will translate business use cases into scalable AI solutions, implement engineering standards for AI application development, and work closely with data, platform, product, risk, security, and business teams to operationalize AI across the organization.

Role Purpose :

The AI Engineer will be responsible for :

- Designing, building, and maintaining production-grade AI applications and agentic systems

- Implementing LLM-based copilots, AI assistants, retrieval-augmented generation (RAG), workflow agents, and autonomous / semi-autonomous agentic use cases

- Utilizing and improving reusable AI engineering patterns

- Following and promoting best practices across prompt engineering, evaluation, guardrails, observability, deployment, and AI application lifecycle management

- Collaborating with cross-functional teams to execute enterprise AI initiatives

Key Responsibilities :

A. AI / GenAI Solution Design & Delivery:

- Build, test, and deploy AI, GenAI, and agentic AI solutions for business use cases

- Implement end-to-end AI solution architectures including the user interaction layer, orchestration / agent framework, LLM inference layer, retrieval layer, memory / context layer, tool / API integration layer, and the monitoring, evaluation, and governance layers

- Develop solutions such as enterprise copilots, internal knowledge assistants, customer support assistants, collections / operations copilots, underwriting / risk / servicing workflow assistants, document intelligence and automated review systems, and agentic workflows for process automation and decision support

B. Agentic AI Architecture & Orchestration:

- Implement single-agent and multi-agent systems for enterprise workflows

- Build agentic patterns such as planner-executor agents, router agents, tool-using agents, workflow agents, retrieval-augmented agents, human-in-the-loop approval flows, and supervisor / specialist multi-agent systems

- Integrate agents with enterprise APIs and microservices, internal knowledge bases, structured and unstructured data, workflow engines, and business rules

- Build robust mechanisms for task decomposition, tool calling, memory / context handling, state management, fallback handling, and retries / exception management

C. Retrieval, Knowledge Systems & Context Engineering:

- Build and maintain RAG / enterprise knowledge retrieval systems

- Develop pipelines for document ingestion, chunking, metadata enrichment, embeddings generation, vector indexing, hybrid retrieval, and reranking

- Optimize relevance and response quality using advanced prompt engineering, retrieval optimization, context assembly, and conversation state / memory strategies

D. AI Application Engineering & Backend Development:

- Build production-grade AI services, APIs, and application backends using modern software engineering practices

- Develop AI microservices and orchestration services in Python and relevant backend frameworks

- Integrate AI systems with enterprise applications, workflow systems, data platforms, and internal APIs

- Ensure systems are modular, testable, secure, and maintainable

E. Evaluation, Guardrails, Safety & Quality:

- Implement evaluation frameworks for AI applications to monitor answer quality, retrieval quality, task completion, hallucination risk, latency, and cost

- Operationalise guardrails for prompt injection resistance, PII / sensitive data handling, response filtering, and policy compliance

- Execute testing approaches for prompts, agent behaviors, tool-use reliability, and benchmark datasets

F. AI Platform, MLOps / LLMOps & Productionisation:

- Support the AI engineering platform and application lifecycle for enterprise AI solutions

- Implement CI/CD and deployment patterns for AI applications

- Follow LLMOps / MLOps practices for versioning (prompts, agents, workflows), evaluation pipelines, telemetry, observability, and cost monitoring

- Work with DevOps / platform teams to deploy AI services reliably across environments

G. Collaboration with Business, Product & Governance Teams:

- Partner with product managers, architects, data teams, risk, compliance, legal, and infosec stakeholders to ensure enterprise readiness

- Contribute to the definition of solution approaches, MVP scopes, and scale-up roadmaps

- Participate in architecture and governance reviews

Required Skills & Experience :

A. Experience :

- 3 to 10 years of experience in software engineering / machine learning engineering / AI engineering / applied AI

- At least 3-4+ years of hands-on experience in building AI / ML / NLP / LLM-based applications

- Strong hands-on experience in GenAI, LLM application development, RAG, and/or agentic AI systems

- Experience deploying applications to production in cloud environments

- Experience working within cross-functional teams alongside product, security, and data stakeholders

B. Core Technical Skills :

1. AI/LLM/GenAI :

- LLM application development and prompt engineering design patterns

- RAG architectures, embeddings, and semantic retrieval

- Vector databases and vector search

- AI agents, tool-using agents, workflow agents, and multi-agent orchestration

- LLM evaluation frameworks and guardrails

- Structured output generation and function / tool calling

- Context management and conversation orchestration

2. Vertex AI / Cloud AI Stack preferred :

- Vertex AI model usage and deployment

- Gemini on Vertex AI

- AI application integration with Google Cloud services

- Model endpoint usage, orchestration, and deployment patterns

- Familiarity with GCP security, IAM, service accounts, and production deployment practices

3. Software Engineering :

- Python (must-have)

- Building APIs and backend services using FastAPI, Flask, or similar frameworks

- Software design, modular architecture, code quality, and unit / integration testing

- Git, CI/CD, containerization (Docker), and deployment pipelines

- REST APIs, microservices, and event-driven integration patterns

4. Data / Retrieval / Search :

- Experience with document processing pipelines and knowledge retrieval systems

- Familiarity with vector stores, search/retrieval pipelines, chunking, and metadata design

- Familiarity with SQL and data access patterns across structured and unstructured data

5. LLMOps / MLOps / Observability :

- Model, prompt, and workflow versioning

- Experiment tracking, evaluation, and logging/tracing

- Monitoring and debugging of AI applications for latency, throughput, and reliability

Location :

BKC, Mumbai