About the Role :
We are looking for a talented and self-driven Backend Engineer with expertise in Python and Node.js to help us build next-generation Agentic AI platforms. In this role, you will design and implement scalable backend systems that power autonomous AI agents capable of integrating with enterprise systems and data platforms. You will work on cutting-edge technologies such as LLMs, multi-agent orchestration, prompt engineering, and real-time adaptive learning systems, collaborating with a cross-functional team of AI researchers, data scientists, and frontend developers.
Key Responsibilities :
- Design, develop, and maintain modular and scalable components for autonomous software agents
- Integrate backend services with modern LLM frameworks such as LangChain, LangGraph, and CrewAI
- Develop APIs, microservices, and orchestration pipelines to manage multi-agent workflows
- Build components for business process automation and CRM integration
- Work with enterprise platforms (e.g., Salesforce, Power Platform, Microsoft Copilot Studio)
- Fine-tune large and small language models (LLMs / SLMs) using proprietary datasets, especially from NBFC domains
- Contribute to model compression, quantization, and performance tuning for edge deployment
- Collaborate with AI teams on model evaluation and optimization pipelines using GCP, BigQuery, and related tools
- Develop lightweight learning frameworks that improve through continuous feedback from user interactions
- Implement real-time models and rules-based decision engines for adaptive automation
- Work on features that allow agents to self-improve over time based on historical performance and results
Required Skills & Experience :
Languages : Proficient in Python, Node.js, with working knowledge of JavaScript, Java, or ReactLangChain, LangGraph, Semantic Kernel, CrewAIAgent Development Kit (ADK), Multi-agent Communication Protocol (MCP)RESTful APIs, WebSockets, Express.js, FastAPI, Microservice architectureGCP, MS Azure, BigQuery, Power Apps, Power BICI / CD tools and containerization (Docker, Kubernetes)Prompt engineering, model orchestration, LLM integrationExposure to fine-tuning, model distillation, and training pipelinesSQL (PostgreSQL / MySQL), NoSQL (MongoDB), vector databases (Pinecone, FAISS)(ref : hirist.tech)