AI Architect - Toa Payoh

AI/Python Architect – Agentic AI Applications

Client: Investment banking

Role Overview: The AI/Python Architect will lead the design and development of advanced AI applications leveraging an agentic approach, where autonomous agents collaborate to solve complex tasks.

This role requires deep expertise in Python, AI frameworks, and multi-agent systems, with a focus on scalable, modular, and intelligent architectures.

Key Responsibilities:

Architect and implement AI systems using agentic principles (e.g., task decomposition, autonomous agents, memory, and tool use).

Design modular Python-based frameworks that support agent collaboration, orchestration, and decision-making.

Integrate LLMs (e.g., GPT, Claude) with external tools, APIs, and databases to enable agent capabilities.

Develop and optimize agent communication protocols, memory systems, and planning strategies.

Ensure scalability, reliability, and security of AI applications in production environments.

Collaborate with data scientists, ML engineers, and product teams to align architecture with business goals.

Evaluate and integrate open-source agentic frameworks (e.g., Google ADK, LangChain, AutoGen, CrewAI, MetaGPT).

Required Skills:

Expert-level Python programming and software architecture.

Strong understanding of AI/ML concepts, especially LLMs and agentic systems.

Experience with frameworks like LangChain, Hugging Face Transformers, Ray, or similar.

Familiarity with cloud platforms (AWS, Azure, GCP) and containerization (Docker, Kubernetes).

Knowledge of prompt engineering, vector databases (e.g., FAISS, Pinecone), and tool chaining.