Senior Generative AI / Applied AI Engineer

Experience: 5+ Years

Location: Hyderabad / Remote – India

Job Type: Full Time

Relocation: Allowed

About the Role

We are looking for a hands-on Senior Generative AI / Applied AI Engineer with strong expertise in LLMs, RAG, AI Evaluation, Context Engineering, Model Optimization, AI Agents, Multimodal AI, and Production AI Systems. The role involves building and deploying scalable AI solutions for a next-generation healthcare platform and clinical assistant.

Key Responsibilities

- Design, develop, deploy, and optimize LLM-powered healthcare applications and AI services using Python.
- Work with OpenAI, Anthropic, LLaMA, Mistral, and other foundation/reasoning models.
- Build production-grade RAG pipelines covering ingestion, document processing, chunking, embeddings, indexing, semantic/hybrid search, retrieval, reranking, grounding, and source attribution.
- Work with vector databases such as FAISS, Pinecone, Qdrant, Weaviate, Milvus, and pgvector.
- Implement prompt engineering, structured outputs, function/tool calling, reasoning workflows, AI agents, and multi-step workflows.
- Build AI evaluation frameworks covering accuracy, relevance, groundedness, hallucination, consistency, safety, retrieval quality, latency, cost, and model stability.
- Implement LLM-as-a-Judge, deterministic/semantic evaluations, retrieval metrics, benchmarking, and human/clinical review.
- Design AI safety, reliability, guardrails, hallucination reduction, prompt-injection protection, PHI/PII protection, validation, fallback, and human escalation mechanisms.
- Develop context engineering, context retrieval, short/long-term memory, patient-context memory, memory retrieval, and summarization capabilities.
- Manage AI releases, regression testing, prompt/model/configuration versioning, A/B testing, canary releases, controlled rollouts, and rollback mechanisms.
- Establish LLMOps and production AI observability, monitoring quality, latency, cost, token usage, retrieval failures, model failures, hallucinations,



safety events, and user feedback.
- Develop multimodal AI solutions involving text, images, PDFs, documents, tables, and structured healthcare data.
- Build and evaluate AI agents, agent orchestration, tool/function calling, reasoning workflows, and human-in-the-loop systems using frameworks such as LangGraph, AutoGen, CrewAI, and LangChain.
- Work with healthcare data including EHR/EMR, clinical notes, patient records, care plans, lab information, medications, diagnoses, CSVs, and clinical knowledge bases.
- Apply healthcare interoperability standards such as FHIR, HL7 and clinical terminologies including SNOMED CT, ICD, and UMLS.
- Deploy and operate production AI systems on AWS, including Bedrock, SageMaker, ECS/EKS, Lambda, S3, RDS/Aurora, OpenSearch, API Gateway, CloudWatch, IAM, and Secrets Manager.
- Work with Docker, Kubernetes, CI/CD, autoscaling, observability, security, high availability, cost optimization, and production troubleshooting.
- Continuously benchmark emerging foundation, reasoning, embedding, reranking, multimodal, and long-context models and recommend production adoption based on quality, reliability, latency, scalability, and cost.

Must-Have Skills
- 5+ years Software Engineering / ML Engineering / Applied AI experience.
- Strong hands-on Python development.
- Production experience building LLM-powered applications and RAG systems.
- Strong knowledge of LLMs, Transformers, Embeddings, RAG, Prompt Engineering, Context Engineering, NLP, Vector Search, and AI Evaluation.
- Experience with commercial/open-source LLMs and vector databases.
- Experience with LangChain and/or LlamaIndex.




- Experience with Hugging Face / Transformers.
- Strong backend/API engineering fundamentals.
- Docker/containerization and cloud deployment experience.
- Experience with AI evaluation, monitoring, Git, and modern software development practices.

Strongly Preferred AWS, Amazon Bedrock, Healthcare/HealthTech, US Healthcare, HIPAA, PHI/PII, EHR/EMR, FHIR, HL7, LLMOps, Kubernetes/EKS, LangGraph, Agentic AI, Multimodal AI, Production AI Observability, Clinical Terminology/Ontologies.

Candidate Profile

Looking for hands-on engineers who can design, code, deploy, evaluate, troubleshoot, and continuously improve production LLM systems. Candidates should understand the trade-offs between quality, reliability, latency, scalability, and cost, and be capable of independently owning production AI systems.

Not suitable for: candidates whose experience is limited to basic ChatGPT/API integrations, simple chatbots, prompt engineering without software engineering, AI POCs without production ownership, or certifications without hands-on production experience.

Interview Expectations

Candidates should be able to explain a production LLM/RAG system they personally built or significantly owned, including architecture, model selection, RAG, chunking, embeddings, vector database, retrieval/reranking, prompts, context engineering, evaluation, hallucination management, guardrails, cloud architecture, deployment, monitoring, scaling, latency, cost, security, and troubleshooting.

Soft Skills

- Strong analytical and problem-solving ability
- Clear written and verbal communication
- Strong documentation
- Independent ownership
- Bias for action with reliability in mind
- Collaboration with Product, Clinical, Engineering, QA, and DevOps teams
- Comfortable working in a fast-paced environment and with global teams

Work Location Hyderabad / Remote | India | Versatile working hours for global collaboration

Remote-friendly with optional Hyderabad base.

📌 Senior Generative AI Engineer (Hyderabad)
🏢 Ekshvaku Tech Innovations
📍 Hyderabad