update: readme

README.md

@@ -11,39 +11,9 @@ pinned: false
 
 > A deep search agentic system for scientific and engineering workflows
 
-**This is a very early work in progress.**
+The objective here is to build a deep search agentic system for scientific and engineering workflows leveraging the power of **Smolagents** and it's coding capabilities. 
 
-## Objectives:
-
-Webcrawler:
-- [x] functions for linkup and tavily
-- [x] function for wikipedia
-- [x] function for arxiv
-- [x] function for pubmed
-- [x] function for sciencedirect
-- [x] function for pdfs
-- [x] function for markdown
-
-Agents:
-- [x] Agent that can work with markdown sources
-- [x] Agent that can search the web 
-- [x] Agent that can analyse big pdfs and markdown sources
-- [ ] Agent that can draw
-- [ ] Agent that can draw in 3D
-- [ ] Agent that can run mechanical simulations
-- [ ] Scientific prompt for engineering tasks
-
-## Deploying:
-- [X] Building a front in React
-- [ ] Connecting the front to supabase for authentification and monitoring credits
-- [ ] Deploying the deepsearch agent on GCP Cloud Run.
-
-And since I am paying, making sure that it is impossible to have an infinite bill.
-
-
-## 🎯 About
-
-DeepEngineer is an open-source project designed to build a specialized agentic system that assists engineers and scientists. The ultimate vision is to create an AI agent that can reason, plan, and execute complex tasks, effectively augmenting and accelerating the engineering and scientific workflow.
+I strongly believe that CodeAgents are superior to answer scientific questions, they only need the right tools to do so.
 
 ### 🚀 Vision
 
@@ -57,9 +27,9 @@ This project aims to tackle this challenge by creating an agent with the followi
 
 ### ⚙️ Current objective
 
-It is likely that this project will not be better than the non specialized deepsearch solutions like OpenAI's or Magnus on scientific questions. Even though those solutions tend to focus on software development, they are generally very good at everything.
+I acknowledge that it is likely that this project will not be better than the non specialized deepsearch solutions like OpenAI's or Magnus on scientific questions. Even though those solutions tend to focus on software development, they are generally very good at everything.
 
-That being said, I hope to learn a lot and I see it as a personal challenge to deploy this solution and then iterate on it. Well see.
+That being said, I hope to learn a lot and I see it as a personal challenge to deploy this solution and then iterate on it.
 
 ## Getting starded
 
@@ -72,7 +42,6 @@ uv sync
 ```
 
 ### 🛠️ Tool Use
-
 The minimum tools needed are:
 - Web crawler agent
 - Scientific paper analyser agent
@@ -85,6 +54,34 @@ The minimum tools needed are:
 
 ### External APIs
 - Mistral (OCR)
+- Mistral (LLM)
 - Deepseek (LLM)
-- Perplexity / Talily / Wikipedia / Arxiv
+- Linkup
+- Tavily
+
+## Roadmap
+
+Webcrawler:
+- [x] functions for linkup and tavily
+- [x] function for wikipedia
+- [x] function for arxiv
+- [x] function for pubmed
+- [x] function for sciencedirect
+- [x] function for pdfs
+- [x] function for markdown
+
+Agents:
+- [x] Agent that can work with markdown sources
+- [x] Agent that can search the web 
+- [x] Agent that can analyse big pdfs and markdown sources
+- [x] Agent that can draw
+- [ ] Agent that can draw in 3D
+- [ ] Agent that can run mechanical simulations
+- [x] Simple Scientific prompt for engineering tasks
+- [ ] Complex scientific prompt in multiple steps for designing engineering systems
+
+## Deploying:
+- [ ] Deploying on huggingface a gradio space
+- [ ] Deploying on vercel a Next.js/Supabase app for authentification and monitoring credits for the front, and GCP Cloud Run for the back.
+- [ ] Deploying the deepsearch agent on GCP Cloud Run.