-
-
The interface to call your future self
-
The interface showing an ongoing conversation
-
The homepage
FutureLens
Talk to Your Future Self. Change Your Present.
Inspiration
Every major decision in life comes with uncertainty.
Should I change careers? Should I move to another country? Will my current habits actually lead me to the life I want?
These questions become even more urgent for people going through major life transitions — such as students choosing a career path or newcomers rebuilding their lives in a new country.
As an international student and newcomer myself, I have often wondered what my life might look like five or ten years from now depending on the choices I make today.
Traditional goal-setting tools give reminders and checklists, but they rarely create an emotional connection with the future.
Psychological research suggests that when people feel connected to their future selves, they make better long-term decisions.
FutureLens was inspired by a simple idea:
What if you could actually talk to your future self?
Instead of imagining the future, FutureLens allows users to have a conversation with AI-generated versions of themselves based on their current habits, goals, and circumstances.
What it Does
FutureLens is an AI-powered life simulation platform that allows users to interact with future versions of themselves.
The experience happens in three phases:
Intelligent Intake Users provide information about their current life situation, goals, habits, and concerns.
AI Simulation Multiple AI agents analyze career trends, financial projections, health patterns, and lifestyle decisions to generate possible future scenarios.
Future Conversation Users can initiate a real-time conversation with their 1-year, 5-year, or 10-year future self and receive insights about how their present choices may shape their future.
After the conversation, the system generates a visual Life Roadmap highlighting key milestones, risks, and decisions that influence long-term outcomes.
How We Built It
FutureLens demonstrates multiple capabilities of Amazon Nova models through Amazon Bedrock.
Frontend
The application interface was built using:
- Next.js 14
- React
- Tailwind CSS
This stack allowed us to create a responsive interface with a smooth onboarding experience and dynamic visual outputs.
AI Reasoning System
FutureLens uses a multi-agent architecture powered by Amazon Nova.
Each agent focuses on a specific domain:
Career Agent Analyzes job market trends and salary trajectories.
Financial Agent Simulates savings growth and cost-of-living projections.
Wellness Agent Evaluates lifestyle and habit patterns.
Lifestyle Agent Considers social environment and life milestones.
Synthesis Agent Combines all outputs into coherent future personas.
These agents are orchestrated using AWS Step Functions, allowing them to run in parallel and combine their insights into a single simulation.
Voice AI
FutureLens enables users to talk directly with their future selves using:
Amazon Nova Sonic
This allows real-time, bidirectional voice conversations between the user and the simulated future persona.
Each future persona has its own system prompt describing:
- personality traits
- emotional tone
- life circumstances
- speaking style
This ensures the conversation remains consistent and believable.
Multimodal Processing
Users can provide information through multiple input types:
- text
- voice narration
- uploaded documents
Amazon Nova processes these inputs into a unified profile before the simulation begins.
Visual Future Roadmap
After the conversation, the system generates a visual roadmap showing possible life trajectories.
This includes:
- timeline projections
- milestone cards
- branching decision paths
- lifestyle visualizations
The roadmap helps users understand how small changes today can affect long-term outcomes.
Modeling Future Outcomes
The simulation combines behavioral patterns with projected outcomes.
For example, a simplified savings model might be represented as:
[ S(t) = S_0 + (I - E)t ]
Where:
- (S(t)) = savings at time (t)
- (S_0) = current savings
- (I) = monthly income
- (E) = monthly expenses
Similarly, habit consistency can influence skill growth over time:
[ Skill_{future} = Skill_{current} + (Learning\ Rate \times Practice\ Time) ]
These simplified projections help the AI construct realistic scenario narratives.
What We Learned
Building FutureLens taught us several important lessons about AI systems.
1. Multi-agent systems produce richer insights
Breaking the simulation into specialized agents significantly improved the depth and realism of the outputs.
2. Prompt engineering is critical
The quality of AI responses depended heavily on well-designed prompts and structured system instructions.
3. Voice interaction creates emotional impact
The most powerful moment of the product experience happens when users hear their future self responding to them in real time.
This transforms the application from a simple prediction tool into a deeply personal experience.
Challenges We Faced
Voice latency
Real-time AI conversations require fast response times. We had to optimize prompts and reduce unnecessary context to keep interactions natural.
Coordinating multiple AI agents
Running several AI agents simultaneously required careful orchestration to ensure consistent outputs.
AWS Step Functions helped coordinate the workflow.
Preventing unrealistic predictions
It was important to ensure the AI did not produce harmful or unrealistic predictions.
We implemented guardrails to ensure the system presents projections and possibilities, not guaranteed outcomes.
What's Next
FutureLens has potential applications far beyond a hackathon prototype.
Possible future developments include:
- career counseling tools for universities
- financial planning assistants
- immigrant settlement guidance platforms
- personal development coaching systems
By combining behavioral science with AI simulation, FutureLens could help people make more informed decisions about their lives.
Final Thoughts
The future is often treated as something distant and abstract.
FutureLens turns it into something you can talk to, question, and learn from.
Because sometimes the best advice we can receive…
is from the person we are becoming.
Log in or sign up for Devpost to join the conversation.