Contextual AI

Build memory and context for AI agents

We integrated memory for RAG 2.0 agents across four layers: working, procedural, semantic, behavioral.

Talk to me about
Cooking

Cook Szechuan, now reinventing Chinese food

Szechuan is my specialty, and I've been learning international cooking to create modern, novel Chinese dishes.

Talk to me about
Contextual AI

Tested whether agent context can self-evolve

It can, with the right setup: detailed LLM self-evaluation as feedback and a strong prior context.

Talk to me about
Software Development Engineer at Amazon·2022 – 2026

Designed context for agentic shopping

I worked on Amazon's context engine and delivered LLM-based customer understanding and recommendations.

Talk to me about
Machine Learning Engineer at X-tech·2020

Cut Nike's China delivery cost 10% with ML

In Beijing I analyzed Nike's distribution system. Earlier I interned in investment research and product data.

Talk to me about
Zhipeng Ye

Zhipeng Ye⁠

About

I build agent memory and cook.

Experiences

  1. Contextual AIBuild memory and context for AI agents

    We integrated memory for RAG 2.0 agents across four layers: working, procedural, semantic, behavioral.

  2. CookingCook Szechuan, now reinventing Chinese food

    Szechuan is my specialty, and I've been learning international cooking to create modern, novel Chinese dishes.

  3. Contextual AITested whether agent context can self-evolve

    It can, with the right setup: detailed LLM self-evaluation as feedback and a strong prior context.

  4. Software Development Engineer at AmazonDesigned context for agentic shopping

    I worked on Amazon's context engine and delivered LLM-based customer understanding and recommendations.

  5. Machine Learning Engineer at X-techCut Nike's China delivery cost 10% with ML

    In Beijing I analyzed Nike's distribution system. Earlier I interned in investment research and product data.

Follow Zhipeng for more...

LinkedIn

© 2026 Delphi · Terms · Privacy

Zhipeng Ye

Zhipeng Ye⁠

About

I build agent memory and cook.

Experiences

  1. Contextual AIBuild memory and context for AI agents

    We integrated memory for RAG 2.0 agents across four layers: working, procedural, semantic, behavioral.

  2. CookingCook Szechuan, now reinventing Chinese food

    Szechuan is my specialty, and I've been learning international cooking to create modern, novel Chinese dishes.

  3. Contextual AITested whether agent context can self-evolve

    It can, with the right setup: detailed LLM self-evaluation as feedback and a strong prior context.

  4. Software Development Engineer at AmazonDesigned context for agentic shopping

    I worked on Amazon's context engine and delivered LLM-based customer understanding and recommendations.

  5. Machine Learning Engineer at X-techCut Nike's China delivery cost 10% with ML

    In Beijing I analyzed Nike's distribution system. Earlier I interned in investment research and product data.

Follow Zhipeng for more...

LinkedIn

© 2026 Delphi · Terms · Privacy

By using this service, you agree to the Terms of Service and Privacy Policy.