How to use Unknown to build social publishing services
A founder developed Linkeme.ai from LinkedIn automation into a multi-platform publishing product, then migrated its workflows to AWS using Cline.
news.ycombinator.comPrompt to try
Editorial adaptation, not the source author’s exact prompt. Replace bracketed inputs and review tool permissions first.
Help migrate [social publishing workflow] from [automation platform] into AWS Lambda and Step Functions services. Model the existing content, ranking, visuals, and publishing stages; implement retry handling and audit trails for [supported platforms].
Reach people exploring AI agents
See placements →What happened
The author reports paying users for version one and an AWS version processing thousands of posts daily with retries and audit trails.
Reported by buzzbyjool on Hacker NewsHow the agent did it
- Scraped article summaries and formatted posts with Make.com.
- Rebuilt the interface in Bolt.new with Make.com backend automation.
- Migrated workflows to AWS Lambda and Step Functions using Cline.
Where a person stepped in
The post does not describe code-review steps; the founder describes successive product iterations and working workflows.
A person approved each step before the agent continued.
Linkeme.ai began with a consulting firm's difficulty maintaining a LinkedIn presence. Its founder first scraped article summaries and formatted posts with Make.com. They then used OpenAI to summarize blogs, create hooks, adjust tone, and incorporate analytics. The author described this early setup as a patchwork without an interface.
The next iteration combined a Bolt.new interface with Make.com backend automation. It integrated GPT-4.1 and Claude for content and semantic understanding, internal post ranking, a platform-specific CTA engine, branded illustration generation, and publishing to LinkedIn, Twitter, Facebook, and Instagram. The author said these workflows worked and attracted paying users.
For version two, the founder moved the product entirely to AWS, using Lambda and Step Functions for scaling. They described Cline as their development tool for LLM agents in VS Code and modeled workflows as AI-defined services. Most Make.com business logic was replaced. The resulting system reportedly processed thousands of posts daily, with retry capabilities and audit trails. The account credits AI with product conceptualization, prototyping, pipeline design, and initial infrastructure code, but does not describe a detailed code-review process.
Who it's for: Developers turning an existing content automation workflow into a publishing product.
Prompt
Suggested prompt (written by Agents In Use, not the author's)
Source: Hacker News · buzzbyjool
Frequently asked questions
Which tools were used for the initial product?
The founder rebuilt the interface with Bolt.new and used Make.com for backend automation, integrating GPT-4.1 and Claude for content and semantic understanding.
What changed in the AWS version?
The author reports moving to Lambda and Step Functions, using Cline for development, replacing most Make.com business logic, and adding retry capabilities and audit trails.
More details
About this account
LLM agents helped conceptualize, prototype, and write infrastructure; publishing capabilities required no human intervention.
Platform: Hacker News. Attribution: https://news.ycombinator.com/item?id=43938227. Prompt: Suggested (written by Agents In Use, not the author's). Sourced by the Agents In Use pipeline.
Credit: buzzbyjool on Hacker News
Source published 9 May 2025
Source checked 14 Oct 2026
Tools and topics
Cline · Bolt.new · Make.com · AWS Lambda · AWS Step Functions
Social publishing · AWS
Editorial completeness: 100/100. Measures documented fields, not agent performance or verified outcomes.