This is a full-time GenAI / Agentic AI Engineer role based in Delhi. The engineer will be responsible for designing, building, and scaling AI-powered products and agentic systems that form the intelligence layer of the Teertham platform.

The role will involve building production-grade LLM applications, AI agents, RAG pipelines, tool-using agents, conversational experiences, recommendation systems, and intelligent automation. You will work closely with product, engineering, design, data, operations, and spiritual/domain experts to transform complex spiritual and customer problems into reliable AI experiences.

A key focus of the role will be building Bhrigu — Teertham’s AI Acharya — and the broader agentic AI ecosystem around it. This includes enabling AI systems to understand scriptures and domain knowledge, reason over structured and unstructured information, use tools and APIs, maintain context, provide personalized guidance, and safely execute multi-step workflows.

The ideal candidate is someone who enjoys working at the cutting edge of GenAI and agentic systems while also caring deeply about reliability, evaluation, latency, cost, safety, and real-world user experience.

Key Responsibilities:

- Design, build, and deploy production-grade GenAI and Agentic AI systems for Teertham’s consumer platform.
- Architect and develop intelligent AI agents capable of reasoning, planning, tool usage, task execution, and multi-step workflows.
- Build and continuously improve Bhrigu, Teertham’s AI Acharya, including its knowledge, reasoning, conversational, and personalization capabilities.
- Develop robust RAG systems over Hindu scriptures, Vedic literature, Panchang, Muhurat, ritual knowledge, and other trusted domain sources.
- Build ingestion, chunking, embedding, retrieval, reranking, citation, and knowledge-management pipelines for domain-specific AI applications.
- Integrate LLMs with internal systems, databases, APIs, search systems, recommendation engines, and third-party tools.




- Build agent tool-calling and orchestration frameworks that enable AI systems to take meaningful actions rather than only generate responses.
- Design conversational AI experiences across chat, voice, and other interfaces where relevant.
- Develop memory and context-management systems to enable personalized and multi-turn interactions.
- Build AI-powered recommendation and personalization systems across pujas, samagri, spiritual services, Panchang, Muhurat, and content.
- Develop intelligent automation for customer support, operations, lead qualification, service discovery, booking workflows, and other business processes.
- Experiment with emerging LLMs, multimodal models, agent frameworks, reasoning models, embedding models, and AI infrastructure.
- Develop prompt engineering, structured-output, function-calling, and model-routing strategies for production applications.
- Build evaluation frameworks to measure factual accuracy, relevance, groundedness, hallucination, safety, task completion, and overall agent quality.
- Create automated and human-in-the-loop evaluation pipelines for GenAI applications.

Establish guardrails to ensure AI responses remain reliable, culturally appropriate, safe, and grounded in trusted sources.
- Monitor AI systems in production for quality, latency, cost, failures, hallucinations, and user behavior.

Optimize inference costs, response latency, model selection, context size, retrieval quality, and overall system performance.
- Build scalable APIs and backend services required to serve AI experiences reliably at production scale.




- Work closely with product managers to translate user problems into AI-powered solutions and determine where AI genuinely creates user value.
- Collaborate with spiritual experts and domain specialists to validate AI outputs and improve the quality of Teertham’s knowledge systems.
- Work with engineering teams to integrate AI capabilities into mobile, web, backend, and operational workflows.
- Stay current with developments in LLMs, agentic AI, RAG, multimodal AI, AI infrastructure, and evaluation techniques.
- Contribute to technical architecture, engineering standards, documentation, and best practices for AI development at Teertham.
- Take ownership from experimentation and prototype through production deployment, monitoring, and continuous improvement.

Experience:

- 3+ years of professional experience in software engineering, machine learning, AI engineering, or a related technical field, with significant hands-on experience building GenAI/LLM applications.
- Strong experience building and deploying production applications using Large Language Models.
- Hands-on experience with agentic AI systems, LLM orchestration, tool calling, function calling, workflow automation, or autonomous/semi-autonomous agents.
- Strong experience with RAG architectures, including embeddings, vector databases, retrieval, reranking, chunking, metadata filtering, and contextual retrieval.
- Experience working with modern LLM APIs and open-source models.
- Experience with frameworks and libraries such as LangChain, LangGraph, LlamaIndex, Semantic Kernel, AutoGen, CrewAI, or equivalent frameworks is preferred.
- Robust programming experience in Python; experience with TypeScript/Node.js is an advantage.
- Strong understanding of REST APIs, microservices, asynchronous processing, databases, and cloud-based architectures.
- Experience with vector databases such as Pinecone, Weaviate, Qdrant, Milvus, pgvector

📌 Agentic AI engineer (India)
🏢 Teertham
📍 India