Ruut Labs
Building AI-Powered MVPs, Apps, Automations systems.
New to Contra
Ruut is building their profile!
Multi-Agent AI & RAG Knowledge System for Real Estate
The challenge: Design a complete agentic AI architecture for a real-estate marketplace — one capable of serving buyers, sellers/brokers, and document-verification workflows through specialized AI agents, grounded in real data rather than free-floating LLM responses.
What we built: A full multi-agent system design comprising a fleet of specialized agents (a buyer-facing conversational concierge, a seller/broker AI co-pilot, a document/listing verification agent, and a pricing-valuation agent), coordinated through a central model-routing and orchestration layer. The system is grounded by a Retrieval-Augmented Generation (RAG) pipeline spanning multiple structured knowledge bases (legal/regulatory corpus, market pricing data, locality/infrastructure data) plus a graph-based knowledge layer linking listings, agents, leads, and locations for relationship-aware retrieval — with human-in-the-loop approval gates on consequential actions.
The outcome: A complete, implementation-ready technical architecture covering agent orchestration, RAG pipeline design (vector + graph retrieval), model-routing strategy, and safety guardrails — the kind of specification an engineering team can build directly from.
1
16
Automated Investor Pitch Deck Generator with Editable Charts.
The challenge: Replace days of manual slide-design work with a repeatable system that generates a fully-branded, investor-ready presentation from structured content - without sacrificing design quality or producing static, non-editable output.
What we built: A script-driven presentation-generation pipeline (Node.js + pptxgenjs) that programmatically assembles a complete, multi-section deck — market sizing, product overview, financial charts, roadmap - applying consistent branding, typography, and layout rules across every slide. Crucially, all charts are generated as native, editable PowerPoint chart objects, not flattened images, so the output deck remains fully editable after generation.
The outcome: An 18-slide, presentation-ready .pptx file generated end-to-end from a single automated build script, with live editable charts and pixel-consistent branding throughout - turning a multi-day manual design task into a repeatable, minutes-long automated process.
1
18
Interactive Physics Web App with Real-Time Canvas Simulation
The challenge: Build a web experience that makes an advanced, abstract physics concept (quantum state superposition and phase-space behavior) genuinely understandable through direct interaction — not a static explainer video.
What we built: A fully responsive, dependency-light HTML/CSS/JS application featuring a live, canvas-rendered phase-space simulation that recomputes and redraws in real time as the user manipulates on-screen controls (a "squeeze" slider and a superposition toggle), plus a second interactive scene simulating a measurement-collapse interaction. The entire experience runs client-side with zero backend, includes bilingual (English/Hindi) content, scroll-triggered reveal animations, and is fully responsive across desktop and mobile.
The outcome: A production-ready, zero-dependency interactive web app demonstrating advanced canvas-based simulation, real-time data visualization, and motion/interaction design — directly applicable to EdTech products, interactive dashboards, scientific/financial simulations, and product demo experiences.
1
24
Multi-Role Interactive Product Prototype: One Codebase, Three Workspaces, Desktop + Mobile
1
1
47