Solo Developer Runs Company with 4 AI Agents on Gemini's Free Tier
Architecture and Implementation
The system uses four AI agents built on OpenClaw (open source), running on WSL2 at home with 25 systemd timers. The agents handle specific tasks:
- Generate 8 social posts daily across platforms with quality gating (generate → self-review → rewrite if score < 7/10)
- Engage with community posts and auto-reply to comments (context-aware, max 2 rounds)
- Research via RSS + HN API + Jina Reader → feed intelligence back into content
- Run UltraProbe (AI security scanner) for lead generation
- Monitor 7 endpoints, flag stale leads, sync customer data
- Auto-post blog articles to Discord when git pushing (0 LLM tokens — uses commit message directly)
Token Optimization Strategy
The developer uses a specific approach to minimize token usage: agents never have long conversations. Every request follows this pattern: (1) read pre-computed intelligence files (local markdown, 0 tokens), (2) one focused prompt with all context injected, (3) one response → parse → act → done. The research pipeline (RSS, HN, web scraping) costs 0 LLM tokens — it's pure HTTP + Jina Reader. The LLM only handles creative/analytical work.
Real Numbers and Infrastructure
- 27 automated Threads accounts, 12K+ followers, 3.3M+ views
- 25 systemd timers, 62 scripts, 19 intelligence files
- RPD utilization: 7% (105/1,500) — 93% headroom left
- Monthly cost: $0 LLM + ~$5 infra (Vercel hobby + Firebase free)
Lessons Learned from Failures
The developer shared specific issues encountered:
- $127 Gemini bill in 7 days: Created an API key from a billing-enabled GCP project instead of AI Studio. Thinking tokens ($3.50/1M) with no rate cap. Lesson: always create keys from AI Studio directly.
- Engagement loop bug: iterated ALL posts instead of top N. Burned 800 RPD in one day and starved everything else.
- Telegram health check called getUpdates, conflicting with the gateway's long-polling. 18 duplicate messages in 3 minutes.
Stack and Resources
Stack: OpenClaw, Gemini 2.5 Flash (free), WSL2/systemd, React/TypeScript/Vite, Vercel, Firebase, Telegram Bot, Resend, Jina Reader. The site (https://ultralab.tw) is fully bilingual (zh-TW/en) with 21 blog posts, with i18n, blog publishing, and Discord notifications all part of the automated pipeline.
GitHub repository with playbook: https://github.com/UltraLabTW/free-tier-agent-fleet
Live agent dashboard: https://ultralab.tw/agent
📖 Read the full source: HN LLM Tools
👀 See Also
Non-developer runs 18-agent OpenClaw setup on Mac mini for digital marketing
A digital marketing agency owner with no coding experience has been running an 18-agent OpenClaw system for six weeks on a Mac mini M4, costing about $100/month for Claude Max Pro plus $5/month electricity. The setup includes three agent households modeled after 'Bridgerton' characters handling content creation, SEO, and development tasks.
Reducing AI Agent Costs by 30% Through Behavior Monitoring and Configuration Changes
A developer cut their OpenClaw bot's token usage by 30% after discovering 70 cron jobs were dumping results into the main chat session, causing context bloat and repeated compaction. The fix involved redirecting cron outputs directly to Telegram and building a monitoring skill to identify inefficiencies like redundant searches and oversized file reads.
Automating Business Vetting with OpenClaw: A Case Study
A developer automated their business verification process using OpenClaw, creating a system that analyzes submitted profiles, checks them against policies, and outputs decisions with reasoning and confidence scores.
Patient uses Claude AI to interpret medical data and navigate brain cancer treatment
A 27-year-old patient with primary mediastinal B-cell lymphoma with CNS involvement uses Claude AI daily to interpret immunohistochemistry panels, analyze PET-CT scan results, evaluate CAR-T clinical trial data, understand drug mechanisms, and prepare questions for medical team.