AI/ML Technical Lead @ Source Desk Global

• AI System Architecture: Design the architecture of the AI-powered agent including LLM-based conversation workflows, voice bots, and follow-up orchestration.
  
• Model Integration & Prompt Engineering: Leverage APIs from OpenAI, Anthropic, or deploy open models (e.g., LLaMA 3, Mistral). Implement effective prompt strategies and retrieval-augmented generation (RAG) pipelines for contextual responses.
  
• Data Pipelines & Knowledge Management: Build secure data pipelines to ingest, embed, and serve tenant-specific knowledge bases (FAQs, scripts, product docs) using vector databases (e.g., Pinecone, Weaviate).
  
• Voice & Text Interfaces: Implement and optimize multimodal agents (text + voice) using ASR (e.g., Whisper), TTS (e.g., Polly), and NLP for automated qualification and call handling.
  
• Conversational Flow Orchestration: Design dynamic, stateful conversations that can take actions (e.g., book meetings, update CRM records) using tools like LangChain, Temporal, or n8n.
  
• Platform Scalability: Ensure models and agent workflows scale across tenants with strong data isolation, caching, and secure API access.
  
• Lead a Cross-Functional Team: Collaborate with backend, frontend, and DevOps engineers to ship intelligent, production-ready features.
  
• Monitoring & Feedback Loops: Define and monitor conversation analytics (drop-offs, booking rates, escalation triggers), and create pipelines to improve AI quality continuously.

Requirements Qualifications
Must-Haves:

• 5+ years of experience in ML/AI, with at least 2 years leading conversational AI or LLM projects.
  
• Strong background in NLP, dialog systems, or voice AI preferably with production experience.
  
• Experience with OpenAI, or open-source LLMs (e.g. LLaMA, Mistral, Falcon) and orchestration tools (LangChain, etc.).
  
• Proficiency with Python and ML frameworks (Hugging Face, PyTorch, TensorFlow).
  
• Experience deploying RAG pipelines, vector DBs (e.g. Pinecone, Weaviate), and managing LLM-agent logic.
  
• Familiarity with voice processing (ASR, TTS, IVR design).
  
• Solid understanding of API-based integration and microservices.
  
• Deep care for data privacy, multi-tenancy security, and ethical AI practices.