Tips on vibe coding

documentation on openAi agents sdk
https://openai.github.io/openai-agents-python/

Day 6
Intro to the OpenAI Agents SDK (a framework for building AI agents)
Tracing which is logging agents requests, inputs, outputs, and assigning a trace title
To run an agent:
-Create an instance of the agent.
-Wrap the execution with with trace() to track it.
-Call Runner.run() to execute the agent.
Started building a Sales Development Agent that automatically sends emails to potential customers on behalf of a company.

#Day6_AE

Day 7

Didn't complete much cuz of mebrat, but we still made progress

Learned agent as a tool — using one agent as a tool that another agent can call to handle tasks.

Learned handoffs — passing responsibility or a task from one agent to another when it's preferred to complete it

#Day7_AE

I need to get out 🫠

about this.... instead of one agent doing everything, agents work together:

✉️ One writes the email.
📝 One creates the subject line.
🎨 One generates a beautiful HTML version.
📤 One sends it.

so you have agents using other agents as tools to send email for your company

tried my multi-agent email system using my personal email as the sender and my other accounts as recipients and it actually worked The agent generates and sends the emails.

...now they're all sitting in spam folder 😂 one more problem to solve

openAI's and anthropics beef is just making my life harder like anthropic doesn't have an openai compatible endpoint all the other models have btw... and I'm using openAI's framework with anthropic's api key even the insturctor recommends not to use them together but that's all i got using them together was a bit tough ...buuut we figured it out anyway

Day 8

-done with sales dev't agent(almost)
-used multiple ai models as multi agents
-saw guardrails(tools to make sure ur agent gets proper info and displays proper info for the user)

#Day8_AE

Good morning my peeps... and we've reached 50 subs🥳🥳 ..... Love y'all🫶

If you're building AI agents, you should get familiar with these 3 common agent/workflow patterns.

Let's break it down.

🔹 Reflection
You give the agent an input.
The agent then "reflects" on its output, and based on feedback, improves and refines.

Ideal tools to use:
- Base model (e.g. GPT-4o)
- Fine-tuned model (to give feedback)
- n8n to set up the agent.

🔹 RAG-based
You give the agent a task.
The agent has the ability to query an external knowledge base to retrieve specific information needed.

Ideal tools to use:
- Vector Database (e.g. Pinecone).
- UI-based RAG (Aidbase is the #1 tool).
- API-based RAG (SourceSync is a new player on the market, highly promising).

🔹 AI Workflow
This is a "traditional" automation workflow that uses AI to carry out subtasks as part of the flow.

Ideal tools to use:
- n8n to handle the workflow.
- GPT-4o, Claude, or other models that can be accessed through API (basic HTTP requests).

If you can master these 3 patterns well, you can solve a very broad range of different problems.

ሸጋ ሰንበት

@naol_builds

heyy y'all... so I didn't get anything done on the agentic course the past two days and decided to try the 4hrs lock-in to catch up and finish the two days videos

just got done with my 4hrs lock-in...well i got distracted a few times🫣, but I've covered two days course

Progress >> perfection 🫡

Day 9 & 10

-Covered Deep Research- using models to search the web and conduct in-depth research for you
-The rest was mostly coding. You'll probably get to see what it is once I deploy it.

Goodnight🌛

#Day9_AE #Day10_AE

So guys we missed after qery sliver project and they came back with an even bigger one Let's not miss this one!!

https://atrium.afterquery.com/r/MG3QTL

Day 11

- deployed my deep research agent(sounds fancy right, it's actually a simple agent), which does multiple web searches, compares the results and displays the best result and sends it as an email

u can try it here if you'd like
https://deep-research-production-e160.up.railway.app/

- that's all I did for today🫠

#Day11_AE

and pls ignore the ui😭