AI Architect
- India
- Permanent
- Full-time
- 10+ years (including 4+ years in GenAI and autonomous agent systems)
- Proven experience designing LLM-powered agents with long-term memory, tool-use, planning, and goal-directed reasoning
- Strong expertise in agentic frameworks like AutoGen, CrewAI, LangGraph, ReAct, CAMEL and BabyAGI
- Experience in multi-agent orchestration, task delegation, and collaborative planning between agents
- Deep knowledge of RAG pipelines, vector search (e.g., FAISS, Weaviate, Pinecone), and embedding strategies
- Strong proficiency in LLM fine-tuning (SFT, LoRA, PEFT) and working with Hugging Face, OpenAI, Claude, etc.
- Skilled in building and integrating tool-augmented agents using APIs, retrieval plugins, code execution, etc.
- Familiarity with LangChain, LangGraph, LangFlow, and other orchestration librarie
- Deep hands-on with Python, FastAPI, cloud services (AWS, GCP, Azure), and infrastructure automation (Terraform, CI/CD)
- Exposure to cognitive architectures (SOAR, ACT-R, OpenCog, Leabra)
- Understanding of agent simulation environments, reinforcement learning, or curriculum-based training
- Experience integrating agents into LMS, enterprise SaaS, or knowledge work platforms
- Knowledge of event-driven, goal-based task scheduling and real-time coordination systems
- Familiarity with Langfuse, PromptLayer, Traceloop, or custom observability pipelines for agents
- Prior work on memory management, context windows, and scalable vector DB partitioning
- Defining the overall architecture of Agentic AI Studio, from agent lifecycle to orchestration and observability
- Designing and deploying scalable agent runtimes, leveraging cloud-native tools and GPU inference platforms
- Leading development of multi-agent coordination protocols, goal resolution, and autonomous workflows
- Driving experimentation with new agentic behaviors, planner-tool integration, and long-term memory systems
- Collaborating with data, infra, and product teams to integrate agentic intelligence into real- world business tasks
- Establishing architectural standards for observability, explainability, and safety of AI agents
- Mentoring engineers and evangelizing agent-based AI thinking across the organization