bebo Technologies - Associate Software Architect - Agentic AI Solutions in Bhubaneswar, भारत

We are looking for an experienced Associate Software Architect - Agentic AI with strong expertise in software architecture, backend engineering, Python development, and emerging AI https://jobeax.com/link/4tHEKzNNbiVhk26b ideal candidate will be responsible for designing, architecting, developing, and deploying scalable, intelligent, and production-ready Agentic AI https://jobeax.com/link/uCbslOSe0BEr08EQ role requires a professional who can combine architectural vision with hands-on coding expertise and lead the development of autonomous AI systems powered by Large Language Models (LLMs), Retrieval-Augmented Generation (RAG), and multi-agent https://jobeax.com/link/TkEUcdyymD55DMj7 candidate should have a strong understanding of distributed systems, microservices, cloud-native applications, and AI engineering, along with the ability to translate complex business requirements into scalable technical https://jobeax.com/link/XnflZD2dGSZYgXFG Responsibilities :- Define and drive the architecture of scalable, secure, and high-performance AI-powered applications.- Design end-to-end software architectures for Agentic AI platforms and intelligent automation solutions.- Develop architectural blueprints, technical specifications, design patterns, and implementation strategies.- Design distributed systems, microservices, APIs, and cloud-native applications.- Ensure system scalability, reliability, maintainability, security, and performance.- Evaluate emerging technologies and recommend appropriate tools, frameworks, and architectural approaches.- Architect and develop autonomous AI agents capable of reasoning, planning, decision-making, and executing complex tasks.- Design and implement multi-agent systems with effective communication, orchestration, and coordination mechanisms.- Develop AI agents using frameworks such as LangChain, LangGraph, CrewAI, AutoGen, or similar technologies.- Implement agent workflows involving tool calling, function calling, memory management, and contextual reasoning.- Build intelligent AI solutions that can independently execute multi-step workflows with appropriate human oversight.- Optimize agent performance, reliability, accuracy, and response quality.- Work extensively with LLMs such as OpenAI GPT, Anthropic Claude, Meta Llama, or other open-source and commercial models.- Design and implement LLM-powered applications for enterprise use cases.- Develop prompt engineering strategies, context management, and model integration techniques.- Implement LLM orchestration, model selection, fine-tuning, and inference optimization where required.- Evaluate model performance, response accuracy, latency, and cost efficiency.- Implement responsible AI practices, guardrails, and output validation mechanisms.- Design and implement scalable RAG architectures for enterprise AI applications.- Develop document ingestion, preprocessing, chunking, embedding generation, and retrieval pipelines.- Integrate vector databases such as Pinecone, Milvus, Weaviate, ChromaDB, or FAISS.- Implement semantic search, hybrid search, reranking, and contextual retrieval techniques.- Optimize retrieval accuracy, relevance, and response quality.- Work with embeddings and knowledge-base integrations to improve LLM responses.- Lead hands-on backend development using Python.- Build scalable backend services, REST APIs, and microservices using FastAPI, Flask, or Django.- Develop reusable, modular, and high-performance software components.- Integrate AI models, databases, third-party services, and enterprise applications.- Implement asynchronous processing, event-driven architecture, and distributed task execution.- Write clean, maintainable, testable, and production-ready code.- Design and deploy AI applications on cloud platforms such as AWS, Azure, or Google Cloud.- Implement CI/CD pipelines, containerization, and orchestration using Docker and Kubernetes.- Establish monitoring, logging, observability, and performance optimization practices.- Build scalable model-serving and inference infrastructure.- Ensure security, data privacy, availability, and fault tolerance across AI applications.- Optimize infrastructure utilization and deployment costs.- Provide architectural guidance and technical leadership to engineering teams.- Mentor developers and establish coding standards and best practices.- Collaborate with product managers, business stakeholders, and engineering teams to understand requirements.- Conduct architecture reviews, code reviews, and technical assessments.- Identify technical risks and recommend effective solutions.- Stay updated with advancements in Agentic AI, Generative AI, LLMs, and software https://jobeax.com/link/fbrGcQB5Clg1ZlqO Experience & Qualifications : - Bachelor's or Master's degree in Computer Science, Information Technology, Engineering, or a related field.- 10+ years of overall experience in software development, backend engineering, software architecture, or AI platform development.- Strong hands-on experience in Python and backend application development.- Proven experience designing and developing scalable microservices-based architectures.- Practical experience building and deploying LLM-based applications and Agentic AI solutions.- Strong understanding of RAG, vector databases, embeddings, and semantic search.- Experience with cloud infrastructure, containerization, and production deployment.- Ability to balance architectural decision-making with hands-on coding and implementation.- Excellent problem-solving, communication, and technical leadership skills.- Experience building enterprise-grade autonomous AI agents.- Knowledge of AI observability, evaluation frameworks, and LLM monitoring.- Experience with LangSmith, Langfuse, or similar tools.- Understanding of model fine-tuning, quantization, and inference optimization.- Knowledge of AI security, prompt injection prevention, and responsible AI practices.- Experience with event-driven architectures, Kafka, Redis, or message queues.- Exposure to MLOps and LLMOps practices (ref:hirist.tech)