Field Notes
Our team explored the thesis of using AI to reconstruct, analyze, and improve decision-making processes. We were interested in understanding how AI can take available information, past decisions, reasoning, and outcomes to provide a clearer reconstruction of how a decision was made and identify areas where the decision-making process could be improved. During the ABC/PVL program, we worked on understanding the problem space, brainstorming possible applications, discussing the feasibility of the idea, refining our approach through mentorship and feedback, and exploring how the concept could be converted into a practical AI-based solution. Timeline: - Initial Phase: Problem identification and brainstorming - Exploration Phase: Research and understanding existing approaches - Validation Phase: Discussing the problem, use cases, and feasibility - Development Phase: Refining the proposed solution and possible implementation - Current Phase: Further exploration and evaluation of the idea EdgeDaemon is a team that enjoys exploring ideas at the intersection of technology and real-world problems. We believe experimentation, questioning assumptions, and learning from failure are as important as the final solution.
Notes & Reflections
“Building ideas that turn complex problems into simple solutions.”
— Vikas N Naik
“Turning ideas into meaningful solutions through technology.”
— Yashas Sadananda
Domains
Margin Note
The ABC program helped us move beyond simply having an idea and taught us to question the problem, validate our assumptions, and think about whether our solution can create real-world impact.
Also in ABC 2026
Vigil
Cross-source signal correlation layer for manufacturing maintenance — correlates alarm logs, sensor trends, and operator observations in real-time to flag developing failures before they happen.
Influex
Creator intelligence platform bridging the gap between creators and brands — data-driven insights, authenticity assessment, portfolio management, and campaign performance tracking.
SpikeNest
Explored industrial sustainability and methane emission management, predictive maintenance using spiking neural networks, and attitude adjustment for CubeSats.
AgentOps
Explored AI-driven agentic solutions across Finance, Real Estate, EdTech, and MSME verticals — focused on identifying meaningful problems and validating real user pain points.
VoxNova
Explored AI reliability across Insurance & claims processing, Customer Service, Geospatial vision, and Warehouse logistics — mapping where AI breaks and how to make it trustworthy.
EquationX
Explored the intersection of Cybersecurity, AI & ML, and Network Security — building intelligent systems that detect and respond to network threats.
Horizon
Startup & problem-solving exploration across ML, Generative AI, and Medical domains — applying Mom Test principles, Reddit community research, and JEPA model exploration to discover real user pain points.