<p><b>Expereince : 3 + Years </b><br/><br/><b>About the Role :</b><br/><br/>We're looking for a Senior AI Developer with deep expertise in building AI agent systems, retrieval-augmented generation (RAG) pipelines, and deploying production-grade GenAI applications.
</p><p><br/></p><p>You will be a core contributor in designing, developing, and scaling intelligent agentic workflows using modern frameworks like LangGraph, CrewAI, and LangChain.
</p><p><br/></p><p>This is a high-impact role for someone passionate about LLMs, context-aware automation, and multi-agent orchestration.<br/><br/><b>Key Responsibilities :</b><br/><br/>- Design and implement AI agent frameworks for task decomposition, tool use, memory handling, and multi-turn conversations.<br/><br/>- Build and optimize RAG pipelines using tools like LangChain, LlamaIndex, or custom vector search setups.<br/><br/>- Integrate agents with internal tools, APIs, and databases to support real-world use cases (e.g., customer support, scheduling, workflow automation).<br/><br/>- Collaborate with ML researchers and product teams to experiment with novel architectures and orchestrators like LangGraph and CrewAI.<br/><br/>- Monitor and evaluate model performance across various use cases using telemetry and custom analytics.<br/><br/>- Ship production-ready systems with robust logging, testing, and monitoring pipelines.<br/><br/>- Stay up-to-date with the latest in LLMs, open-source agentic frameworks, and vector search infrastructure<br/><br/><b>Required Skills & Qualifications :</b><br/><br/><b>Must-Have Experience :</b><br/><br/>- Programming - Python (advanced), Typescript/Node.js (nice to have)<br/><br/>- AI Frameworks : LangGraph, CrewAI, LangChain, LlamaIndex, OpenAI, Hugging Face<br/><br/>- Agent Systems : Designing task-oriented agents with memory, tool use, planning, and inter-agent communication<br/><br/>- RAG Architecture : Document loaders, chunking strategies, vector embedding models, hybrid search (BM25 + vector), contextual reranking<br/><br/>- LLM Tooling : OpenAI GPT-4/4o, Claude, Gemini, local models (e.g., Mistral, LLaMA)<br/><br/>- Infrastructure : Vector DBs (e.g., Weaviate, Pinecone, Qdrant, Elasticsearch), Postgres, MongoDB<br/><br/>- MLOps : Prompt engineering, model evaluation, A/B testing, telemetry, observability<br/><br/>- Deployment : REST APIs, FastAPI, Docker, CI/CD pipelines<br/><br/>- Other : Strong written and verbal communication; ability to work independently and own initiatives end to end</p> (ref:hirist.tech)