Inspiration
The Immigration Landscape shifts every day, affecting a lot of immigrants. It is not possible to predict the future and prepare a stable roadmap ensuring immigration success, but it is possible to prepare for worst case scenarios!
What it does
We generate a realistic immigration plan, and find loop holes that could drasticlly derail the plan and build backup plans an immigrant could follow to save guard his chances for successful immigration into the US.
How we built it
We gave the following Inputs to the planner, for a proof of concept: 1. Your immigration goal is to be achieved in the future 2. Current status/ position 3. Resume (optional) (pdf parsed and formatted by an LLM agent) 4. Travel history (optional) (pdf parsed and formatted by an LLM agent) 5. Additional information (optional text)
Output by the AI Planner: 1. A realistic plan (road map) to achieve your immigration goal. #internally uses a RAG knowledge base to get real info on visa wait times countrywise, OPT info and Naturalization requirements. 2. wild scenarios that you could face by following the given plan. 3. contingency plans that can be followed if any of the wild scenarios occur. All in a compact PDF!
Challenges we ran into
- Lots of Inputs that can drastically affect the plans, but we reduced it Resume, Travel History to show a proof of concept
- LLM Token Limits.
- Slow running time ## Accomplishments that we're proud of We achieved all our goals , we added a RAG, Parsed PDF inputs, Generated a great Output PDF We learnt new stuff as we did the project
What we learned
- RAG Integration
- Gradio Webpage Design
- Parsing and Formatting PDFs ## What's next for Immigration Roadmap Generator
- Scaling Up the Inputs, LLMs
- Expanding the RAG Database.
Built With
- claude
- gradio
- langchain
Log in or sign up for Devpost to join the conversation.