How to build your own AI agent:
AI agents are one of the biggest growth areas in 2025.
Understanding AI agents will be one of the most valuable skills of the year.
So how do we create one?
First, AI agents typically fall into two categories:
Those without memory — stateless agents.
They react only to immediate input, like a blank slate every time.
Those with memory — stateful agents.
They leverage past interactions to deliver context-aware, smarter, and highly personalized responses—making them much more powerful and capable of highly complex workflows.
Memory is where the magic happens.
Here's a great read on memory: https://t.co/2Gy4AvbnkU
The article above by @metaskills and @zep_ai provides a fantastic explanation on:
🔹 How memory works
🔹 How to integrate memory into an AI agent
Now let’s walk through the key steps to build an AI agent with memory:
1) Define the AI agent's purpose
Start by clarifying its role, such as a personal assistant or customer service bot, and determine the data it needs to process (eg; user profiles or task history).
2) Choose LLM models
Select an LLM or LLMs (e.g; OpenAI, Hugging Face) that match your requirements.
3) Plan memory requirements
Design for both short-term memory (conversation context) and long-term memory (persistent user knowledge).
4) Integrate Zep for memory management
Use Zep’s framework to manage memory, sessions, and knowledge graphs.
5) Build user profiles and sessions
Create user IDs and maintain session continuity for seamless interactions.
6) Design a workflow
Establish processes for memory management. Define workflows to retrieve, update, and use memory for informed responses.
7) Incorporate a knowledge graph
Use Zep's knowledge graph to store and query relationships, and enrich responses.
8) Develop and test contextual prompts
Create prompts that dynamically use memory and knowledge graphs for relevant, secure responses.
9) Secure data
Implement robust security measures and restrict data access.
10) Monitor, improve, and scale
Use analytics and feedback to refine memory systems and scale capabilities.
These steps provide a high-level guide for building stateful AI agents.
If you want to dive deeper into how memory works or integrating it, here's a great article: https://t.co/2Gy4AvbnkU
💭 Over to you. Have you worked on projects with AI agents? 💬
This evening I’ve been hosted on Radio for the first time.💃
I’ve been on @SpiceFMHoima talking the Digital Rights for Girls and Women Project by @EnabelinUganda that envisions economically empowered women through digital access and the ability to claim their digital rights.
“The decisions you take in law school today will determine the spaces you’ll be invited to in the future”-@jojo_luzige
The event may be done, but the impact & memories will always last!
All thanks to the incredible panelists, lots of insights were taken from this.
#EventLegacy
arXiv -> alphaXiv
Students at Stanford have built alphaXiv, an open discussion forum for arXiv papers. @askalphaxiv
You can post questions and comments directly on top of any arXiv paper by changing arXiv to alphaXiv in any URL!
@DoleYoMwana@mungomaKevin@nyombiluzze AI didn't spring up last year but most these old fellows have no experience or good understanding of the current State of the Art models and architectures.