Background
AI is becoming central to how we work — yet current tools fall short. They force people to work around them instead of working for the user. HuKu was born from a vision:
A living, intelligent environment where human and AI agents collaborate naturally to automate routine tasks, accelerate strategic work, and unlock new levels of potential.
The Problem
Four core barriers limit current LLMs in the workplace:
- Knowledge Silos — company data trapped in PDFs, Slack, emails
- Poor UX — AI cannot handle complex, multi-step work
- Single-Model Limits — specialized tasks need specialized intelligence
- Workflow Isolation — AI disconnected from actual work applications
AI feels like texting an intern who forgets everything between conversations.
The Solution
HuKu is a multi-LLM agentic workspace with built-in intelligence:
- Multi-LLM system — orchestrating OpenAI, Anthropic, and Gemini
- AI Knowledge Base — a "Second Brain" that retains context
- AI-Native Work Interfaces — Docs, Sheets, and Canvas built for AI collaboration
- Expert Agents — Sales, Support, Ops, and Marketing specialists
- No-Code Deployment — customizable without engineering
- Deep App Integrations — Slack, Gmail, Drive, CRMs
- Memory + Tool Access — persistent context across sessions
My Role & Approach
Research Phase
Studied hundreds of AI products, conducted user interviews, and identified critical gaps around fragmented knowledge and isolated AI tools.
Design Pillars
- Multi-LLM Orchestration
- Second Brain Knowledge Layer
- AI-Native Work Interfaces
- Agentic Workflows
What We Shipped
- Collaboration within Docs, Sheets, and Canvas
- Persistent agent memory across sessions
- Cross-tool integration (Slack, Gmail, Drive, CRMs)
- No-code customization for non-technical users
- Dynamic model routing across LLM providers