Job Summary
We are seeking a skilled Python developer to design, develop, and deploy AI-powered Microsoft Teams Bot Applications using the Microsoft 365 Agent SDK. The role involves integrating Large Language Models (LLMs) with Azure AI Search (and other vector databases) to implement Retrieval-Augmented Generation (RAG) solutions. The ideal candidate will have a strong background in Python, Microsoft Bot Framework, Azure AI services, and modern AI/ML techniques to create intelligent, scalable, and user-friendly bot experiences.
Key Responsibilities
1. M365 Agent SDK & Bot Development
- Design, develop, and deployMicrosoft Teams Bot Applicationsusing theM365 Agent SDKandPython.
- Implementconversational AIworkflows, including natural language understanding (NLU), dialog management, and multi-turn conversations.
- Integrate bots withMicrosoft Teams, Outlook, and other M365 servicesusing Microsoft Graph API.
- Optimize bot performance forscalability, latency, and user experience.
2. AI & LLM Integration
- Develop and fine-tuneLLM-based solutions(e.g., Azure OpenAI, Hugging Face, or custom models) for conversational AI.
- Implementprompt engineeringandcontext managementto enhance bot responses.
- IntegrateAzure AI Search(or other vector databases like Pinecone, Weaviate, or Chroma) forsemantic search and RAG.
- Design and optimizeembedding pipelinesfor document retrieval and knowledge base augmentation.
3. RAG & Vector Database Implementation
- BuildRetrieval-Augmented Generation (RAG)systems to ground LLM responses in enterprise data.
- Index, chunk, and embedunstructured data(PDFs, documents, emails) for efficient retrieval.
- Evaluate and comparevector databases(Azure AI Search, Pinecone, Weaviate, etc.) for performance, cost, and scalability.
- Implementhybrid search(keyword + vector) for improved accuracy.
4. Cloud & DevOps
- Deploy and manageAzure resources(App Services, Container Instances, etc.) for bot hosting.
- Implement/modifyCI/CD pipelinesfor automated testing and deployment.
- Ensuresecurity best practices(authentication, authorization, data encryption) for bot and AI services.
5. Collaboration & Documentation
- Work closely withcross-functional teams(product managers, UX designers, data scientists) to refine bot capabilities.
- DocumentAPIs, workflows, and system designfor internal knowledge sharing.
- Providetraining and supportto end-users and stakeholders.