Have you ever felt frustrated with the limitations of existing research tools? I know I have. As someone who relies heavily on deep research for my work, I’ve found that tools from OpenAI and Perplexity often fall short. The output is hard to control, and I’m left waiting for 15 minutes or more to see if my prompt is on the right track.
I think the root of the problem lies in how these models are trained. They’re based on data produced by trained annotators, not my own research style or framework. So, I decided to take matters into my own hands and build a custom research agent tailored to my needs.
Using an open-source framework and calling Gemini underneath, I created a tool that’s designed to help me dig deeper and faster. Here are some of the key features that make it stand out:
## Prompt Improvement
I added a clarifying question step to help refine my prompts and get more accurate results.
## Editable Search Plan
Before starting my research, I can modify my search plan to ensure I’m on the right track.
## Step-by-Step Execution
The tool executes my search plan step-by-step, automatically pivoting or extending directions as results come in.
## Super Deep Research
This is where the magic happens. I can dive as deep as 10+ steps, with 10+ queries in each step, to uncover insights that might have otherwise gone unnoticed.
I’d love to share this tool with the community and hear your feedback! Check it out at https://myintelliagent.com/.
What do you think? Have you struggled with the limitations of existing research tools? Share your experiences and suggestions in the comments below!