Job Description
We are a technology-led healthcare solutions provider. We are driven by our purpose to enable healthcare organizations to be future-ready. We offer accelerated, global growth opportunities for talent that’s bold, industrious, and nimble. With Indegene, you gain a unique career experience that celebrates entrepreneurship and is guided by passion, innovation, collaboration, and empathy. To explore exciting opportunities at the convergence of healthcare and technology, check out www.careers.indegene.com Looking to jump-start your career? We understand how important the first few years of your career are, which create the foundation of your entire professional journey. At Indegene, we promise you a differentiated career experience. You will not only work at the exciting intersection of healthcare and technology but also will be mentored by some of the most brilliant minds in the industry. We are offering a global fast-track career where you can grow along with Indegene’s high-speed growth. We are purpose-driven. We enable healthcare organizations to be future ready and our customer obsession is our driving force. We ensure that our customers achieve what they truly want. We are bold in our actions, nimble in our decision-making, and industrious in the way we work.
We are looking for a highly skilled and forward-thinking AI Solutions Architect to lead the design, development, and deployment of enterprise-grade AI solutions powered by Large Language Models (LLMs), Agentic AI, and Retrieval-Augmented Generation (RAG). This role demands hands on expertise in designing scalable AI architectures, integrating cutting-edge tools, and ensuring safety, observability, and performance at scale. This role also requires collaboration with engineering leads to drive AI innovation/ discussion, drive technical decisions, ensure delivery ownerships and include Gen AI capabilities in the existing product roadmaps/ frameworks.
The ideal candidate brings strong experience in Software engineering, AI/ML systems, system/ solution engineering, distributed architecture, and platform design, along with recent hands-on work in Generative AI, multi-agent systems, LAG/RAG pipelines, and AI governance. You should be equally comfortable reviewing legacy software architectures, mentoring engineers, working with cross-functional stakeholders, and driving production readiness of AI systems.
Must Have
Key Responsibilities :-
AI Platform Architecture: Design and own the end-to-end architecture for enterprise AI systems spanning LLM orchestration, multi-agent workflows, RAG pipelines, vector databases, LLM gateways, and API layers. Define reference architectures, design patterns (ReAct, Plan-and-Execute, Tool-Use), and enforce architectural standards through Architecture Decision Records (ADRs).
Agentic AI & Multi-Agent Systems: Architect and implement production multi-agent systems with specialized agents (content generation, compliance auditing, data retrieval, orchestration), inter-agent communication protocols, tool/function calling frameworks, plugin registries, and memory management (conversation buffers, working memory, long-term vector stores). Design human-in-the-loop checkpoints, fallback chains, and graceful degradation patterns.
Production RAG Systems: Design production-grade RAG pipelines including document ingestion (PDF, DOCX, HTML with tables, images, multi-column layouts), chunking strategies (recursive, semantic, document-structure-aware, parent-child), hybrid retrieval (semantic + BM25 + metadata filtering), re-ranking, query analysis/rewriting, citation enforcement, and hallucination detection.
LLM Strategy & Model Routing: Define multi-provider LLM strategies across Claude (Anthropic), GPT-4o (OpenAI), Gemini (Google), and open-source models (Llama, Mistral). Architect intelligent model routing that optimizes for cost, latency, and capability. Evaluate and recommend models for specific use cases based on capability fit, context window, compliance, and deployment constraints.
LLMOps & AgentOps: Build CI/CD pipelines for AI systems prompt versioning and regression testing, automated evaluation suites (RAGAS, LLM-as-Judge), model A/B testing, and staged rollouts. Design observability infrastructure with full execution tracing, token tracking, cost attribution, latency SLAs, and quality dashboards using AgentOps, LangSmith, Langfuse or OpenTelemetry.
Evaluation & Quality Assurance: Design multi-layered evaluation frameworks combining automated metrics (RAGAS, BLEU, ROUGE), LLM-as-Judge pipelines with bias mitigation, embedding-based similarity, and domain-expert human review. Build and maintain golden test datasets (500+ curated QA pairs). Define monitoring for retrieval health, hallucination rates, empty retrieval rates, and user satisfaction.
Explainable AI (XAI): Implement explainability at multiple levels input attribution (RAG citations with chunk-level source mapping), reasoning traces (structured Chain-of-Thought with audit trails), confidence calibration, uncertainty quantification, and contrastive explanations. Produce model cards, data sheets, and system documentation for regulatory compliance.
Responsible AI & Governance: Establish and lead the Responsible AI program including guardrail architecture (input/processing/output guardrails with declarative configuration), bias detection and fairness evaluation pipelines, red-teaming exercises (prompt injection, data poisoning, adversarial attacks), AI security measures, PII redaction, privacy-by-design, and regulatory compliance (EU AI Act, NIST AI RMF, FDA AI/ML guidance, GDPR/CCPA).
Strategic Leadership: Collaborate with Engineering, Product, Business, Legal, and Compliance stakeholders to translate requirements into AI roadmaps. Mentor a team of 20+ engineers through design reviews, technical coaching, and career development. Drive build-vs-buy-vs-open-source decisions and present technical strategy to executive leadership.
Good to have
Technical Skills :-
Languages & Frameworks: Expert Python; proficient JavaScript/TypeScript. FastAPI, Flask, or Django for backend APIs. ReactJS or Streamlit for AI interfaces.
LLM & Agent Frameworks: Microsoft Agent framework, AutoGen, LangGraph, LangChain, CrewAI, Strands Agents, Semantic Kernel. Deep knowledge of ReAct, Plan-and-Execute, and Tool-Use orchestration patterns.
Cloud & Infrastructure: AWS (Bedrock, SageMaker, Lambda, S3, DynamoDB) and/or Azure (Azure OpenAI, AI Studio, Cognitive Services). Multi-cloud experience preferred.
Vector Databases: Pinecone, Weaviate, Qdrant, pgvector, ChromaDB, or Milvus in production environments.
Document Processing: Azure Document Intelligence, PyMuPDF, Unstructured.io, LlamaParse. Experience with tables, images, and complex PDF/DOCX extraction.
Data & Storage: MongoDB, PostgreSQL, Redis, Elasticsearch. Data modeling for AI-driven applications.
Observability & MLOps: AgentOps, LangSmith, Arize Phoenix, OpenTelemetry, Weights & Biases. CI/CD for AI (GitHub Actions, Jenkins).
LLM Gateway: LiteLLM, Portkey, or custom gateway development for multi-model routing, rate limiting, and cost tracking
Key Competencies :-
Strong architectural thinking with the ability to balance scalability, simplicity, cost, speed, and risk.
Strong people leadership and mentoring capability.
Ability to drive delivery while maintaining technical quality.
Excellent communication and stakeholder management skills.
Strong problem-solving and decision-making ability in ambiguous environments.
High ownership mindset with focus on business impact and production outcomes.
Commitment to responsible, secure, and ethical AI adoption.
Education :-
Bachelor’s or Master’s degree in Computer Science/ Artificial Intelligence, Machine Learning, or a related quantitative field.
Ph.D./ Publications or contributions to AI projects/ international journals is a plus