FlowLens
Inspiration
AI adoption is moving faster than ever, but many organisations struggle with the most important first step: understanding where AI can genuinely improve their business.
We were inspired by the idea that AI should not simply replace processes or automate tasks blindly. Instead, it should act as an intelligent assistant that helps people discover opportunities, make better decisions, and transform the way they work.
FlowLens was created to bridge the gap between emerging AI capabilities and real business value, helping teams move from "Where can we use AI?" to "Here is where AI can make a measurable difference."
What it does
FlowLens allows users to upload business documentation such as:
- Process documents
- Standard operating procedures
- Meeting notes
- Requirements documents
- Workflow descriptions
The AI analyses the information and generates:
- Current-state process maps
- Key stakeholders and systems involved
- Bottlenecks and inefficiencies
- AI and automation opportunities
- Business impact analysis
- User stories and acceptance criteria
- Implementation roadmap
Users can then interact with the analysis conversationally, asking questions such as:
- "Which steps are best suited for AI assistance?"
- "What risks should we consider?"
- "How could this process be improved?"
FlowLens transforms AI adoption from an abstract idea into a structured, actionable plan.
How we built it
FlowLens was built as a modern full-stack AI application.
Frontend
- React
- TypeScript
- Tailwind CSS
Backend
- Python
- FastAPI
- PostgreSQL
AI Layer
- OpenAI models
- Agentic workflow design
- Structured outputs for reliable analysis
The system was designed around specialised AI capabilities rather than a single prompt. Different analysis stages handle process understanding, opportunity discovery, business recommendations, and implementation planning.
The architecture was designed to be modular, allowing future expansion into more autonomous AI agents and enterprise workflows.
Challenges we ran into
One of the biggest challenges was ensuring AI recommendations were practical rather than generic.
A simple AI response can describe possibilities, but a useful business tool needs context, structure, and reasoning.
We focused on:
- Creating reliable structured outputs
- Designing effective AI workflows
- Balancing automation with human decision-making
- Converting unstructured business information into actionable insights
- Building a user experience that feels like a professional consulting tool
Accomplishments that we're proud of
We are proud of building a working AI transformation assistant that goes beyond simple chatbot interactions.
Key achievements:
- Created an end-to-end workflow from document upload to transformation recommendations
- Designed an agentic architecture for future expansion
- Combined software engineering practices with modern AI capabilities
- Built a tool focused on real organisational challenges rather than technology experimentation alone
What we learned
We learned that successful AI adoption is about understanding workflows, people, and business objectives.
The most valuable AI solutions are not necessarily the ones that automate the most. They are the ones that help humans make better decisions and improve how organisations operate.
We also learned the importance of designing AI systems with clear boundaries, structured outputs, and human oversight.
What's next for FlowLens
Future development will focus on making FlowLens a complete AI consulting assistant.
Potential next steps:
- Real-time workshop analysis during client discovery sessions
- Integration with project management platforms
- Automatic generation of technical architectures
- ROI estimation for AI initiatives
- Enterprise knowledge integration
- More autonomous AI agents for research, planning, and implementation support
Our long-term vision is for FlowLens to become a trusted AI partner that helps organisations navigate technological change and adopt AI responsibly.
Log in or sign up for Devpost to join the conversation.