Join Our Team as a Lead AI Architect
Kandou is seeking a Lead AI Architect to design, build, evaluate, and deploy advanced AI agent systems for real-world applications. This role focuses on innovative agentic systems that extend beyond conversational assistants to include complex analytical workflows, knowledge-based reasoning systems, controlled inference pipelines, tool-using agents, and transparent decision-support architectures.
About the Role
This is an opportunity for someone who excels in transforming research theories into actionable systems. Key activities include designing agent architectures, implementing reasoning workflows, testing system reliability, creating user-facing interfaces, and ensuring agentic behavior is interpretable, controllable, and robust.
Key Responsibilities
- Design and implement real-world agentic AI systems using modern agent frameworks and orchestration tools.
- Develop agentic workflows that extend beyond chat, including:
- Complex analytical pipelines
- Multi-step research workflows
- Tool-using agents
- Knowledge-grounded agents
- Structured decision-support systems
- Work with knowledge-based AI architectures, incorporating:
- Retrieval-augmented generation
- Knowledge graphs
- Symbolic rules
- Structured domain models
- Ontologies
- Hybrid reasoning systems
- Develop mechanisms for controlling inference, including planning constraints, reasoning policies, guardrails, and validation layers.
- Implement neuro-symbolic approaches for agentic reasoning by merging LLM-based reasoning with symbolic and rule-based methods.
- Create transparent AI methodologies that enhance the traceability and explainability of agent behavior.
- Establish evaluation frameworks for agentic systems encompassing benchmarking tasks, regression tests, and robustness testing.
- Develop full-stack prototypes and production applications, integrating backend services, APIs, and frontend interfaces.
- Collaborate with researchers, engineers, and domain experts to convert complex real-world challenges into effective agentic workflows.
- Stay updated on advancements in agentic AI, reasoning systems, and applied neuro-symbolic methods.
Required Experience
- Extensive experience in developing real-world AI agents or workflows.
- Focus on agentic reasoning, including planning, tool usage, and autonomous task completion.
- Experience in industrial AI development, academic research, or both.
- Hands-on familiarity with knowledge-based agentic systems.
- Knowledge of techniques for controlling reasoning or inference.
- Familiarity with neuro-symbolic AI concepts.
- Experience with evaluation and benchmarking of agentic systems.
- Full-stack web development skills.
Technical Skills
- Solid Python engineering skills.
- Experience with modern LLM and agentic AI frameworks such as:
- LangChain
- LangGraph
- OpenAI SDK / OpenAI Agents SDK
- Tool/function calling
- Backend development knowledge, including APIs and databases.
- Familiarity with frontend technologies like React, Next.js, or TypeScript.
- Knowledge of vector databases, graph databases, and knowledge graph tooling.
Qualifications and Portfolio
- Significant open-source contributions.
- Portfolio projects related to agentic AI and LLM systems.
- Experience with AI systems in domains like scientific analysis or decision support.
- Production-grade AI system design experience, including security and reliability.
- Knowledge of human-in-the-loop systems and workflow auditability.
Apply online using the form below. Only applications matching the job profile will be considered.