← Back to portfolio

Selected work

GTMage

GTMage combines conversation discovery, AI-assisted reply drafting and human review in a desktop social-content workflow.

Product walkthrough

The problem

Finding a relevant conversation is only the beginning of useful outreach. The response also needs to fit the topic, the people involved and the author’s voice. Moving between social feeds, research and a separate drafting tool can make that context hard to retain. The product problem was connecting discovery with thoughtful preparation while keeping the author involved in the final wording. A system that produces a draft still needs a clear review step before that text becomes a response on a public platform.

What was built

I designed and built GTMage as a desktop social-content workflow. It discovers conversations across X, LinkedIn, Reddit and Product Hunt through feeds, topic monitoring and web search. Users can draft responses with a choice of AI models and an author voice, then review and edit the wording. Research, drafting and review stay together in the application. The workflow can prepare a response for the target platform by prefilling the reviewed text. Credentials are stored in the system keychain, keeping account access within the desktop setup.

Result or current status

The portfolio walkthrough presents the product’s discovery and drafting experience. My contribution covers both the product design and the implementation of the social-content workflow. The demonstrated result is a way to move from a relevant conversation to an editable response with its context still available. Human review is part of that route. The author can inspect the research, adjust the draft and decide how to use the response on the target platform. That review experience is central to the product I built: it keeps the conversation and proposed wording together, helping the user prepare a reply while retaining control of the text that represents them.

Play nowStep insideWalk the Pavilion. Play chess or ping-pong.