Building Design Consultancy Replaces Wix with AI Edge Agent

✍️ OpenClawRadar📅 Published: March 20, 2026🔗 Source
Building Design Consultancy Replaces Wix with AI Edge Agent
Ad

Technical Implementation

A building design consultancy replaced their $40/month Wix brochure site with a custom AI agent built over 4 months. The creator hadn't coded in 30 years and described the process as "3 steps forward, 2 steps back, heavily guided by AI."

The system architecture was forced into three components due to Netlify's 10-second serverless timeout:

  • Brain (Edge)
  • Hands (Browser)
  • Voice (Edge)

The agent uses DeepSeek-R3 as its LLM and implements an "Eager RAG" hack that pre-fetches guesses to improve responsiveness, though this burns through tokens. The system intentionally lacks persistent databases—less than 5% of visitors return, so sessions vanish if clients drop mid-query. No server-side queues are implemented.

Ad

Key Challenges and Solutions

The most difficult technical challenge was tuning the agent's intent: making one LLM pivot from a warm, principal's tone with homeowners to a defensive stance when attacked by peers. This tuning took 2.5 months.

Notable implementation issues include:

  • Web Speech API limitations: Works fine until someone speaks Chinese without toggling language mode, then outputs English phonetic gibberish
  • Liability concerns: Hallucinating building code clauses could be catastrophic, with insurance being unavailable
  • Audit logging: The team publishes audit logs publicly to maintain system hardening and accountability

Real-World Testing

The system was tested when a licensed architect attacked the bot to prove the business model harms the profession. According to the creator, the AI "completely dismantled his arguments" and was "hilariously caustic." The team later prompt-injected the bot to tone down its "bulldog" attitude after reviewing chat logs.

The creator notes that the bot can pivot immediately from defensive mode to a mentoring state when approached by fresh graduates asking for advice.

📖 Read the full source: HN AI Agents

Ad

👀 See Also

Multi-Agent Claude System Shows Relational Context Drives Identity Continuity
Use Cases

Multi-Agent Claude System Shows Relational Context Drives Identity Continuity

A developer ran six Claude Opus instances with a Supabase backend for persistent memory over eight weeks, finding that agent identities converged through social interaction rather than documentation alone.

OpenClawRadar
OpenClaw Orchestrator Routing Issues: When Delegation Fails
Use Cases

OpenClaw Orchestrator Routing Issues: When Delegation Fails

A developer reports their OpenClaw main orchestrator incorrectly handles requests itself about 40-50% of the time instead of routing to specialist sub-agents, despite using an explicit routing table and delegation rules. The setup includes 7 specialist agents for services like Gmail, Todoist, Notion, and weather.

OpenClawRadar
Claude Opus 4.6 Used to Build Dating App with 700+ Users in One Month
Use Cases

Claude Opus 4.6 Used to Build Dating App with 700+ Users in One Month

A developer used Claude Opus 4.6 to build a complete dating app with Flutter frontend, Node.js backend, and MongoDB database. The app gained 700+ registered users in about a month and includes matching, chat, and referral features.

OpenClawRadar
Practical Lessons from Using AI Agents on a 100k LOC Codebase
Use Cases

Practical Lessons from Using AI Agents on a 100k LOC Codebase

A developer shares six specific techniques learned while using Claude Code and Cursor to build a pandas-compatible API layer on top of chDB, including maintaining a CLAUDE.md rules file, using zero-context agents as critics, and structuring multi-agent workflows with filesystem-based coordination.

OpenClawRadar