The future of personal reputation may not rely solely on resumes, job titles, social media posts, or self-promotion.
As AI increasingly becomes an intermediary for searching, matching, recommending, collaboration, and dispute resolution, knowledge workers — including lawyers, researchers, developers, and consultants — face a new challenge: our real value often lies hidden in tedious documents, draft iterations, and complex debugging processes. This kind of work is difficult to display publicly, hard for AI systems to recognize, and challenging for others to quantify.
To explore a possible solution, I developed and open-sourced VeriHash, an experimental tool designed to preserve a long-term, machine-readable footprint of professional work.
Core Positioning
VeriHash does not prove that a person is excellent.
It simply helps preserve a long-term, machine-readable trail of what that person has actually worked on.
How It Works
VeriHash acts like a personal “digital notary.” When a task is completed, it can:
-
AI Smart Extraction
Locally distill work increments and generate sanitized workload summaries and skill tags. -
Cryptographic Anchoring
Timestamp the work and generate a unique credential ID. -
Web3 Identity System
Each user generates a local DID (Decentralized Identifier) key. This globally unique identity key applies a digital signature to each work record. -
Hash Linking
After signing, each new credential is mathematically linked to the hash of the previous one, forming a tamper-evident hash chain.
Ultimately, this data is packaged into a lightweight personal Credential Chain.
VeriHash’s core value lies in the long-term continuity of one’s work footprint and identity ownership.
Why Broadcast via GitHub Gist?
Once a credential chain is built, AI systems need a stable, machine-readable source to discover it.
We chose GitHub Gist as the public broadcasting channel because it provides stable URLs and plain-text, machine-readable records. This allows AI systems to read signed “intellectual fingerprints” directly, without relying solely on third-party intermediaries.
Latest Milestone
VeriHash has just released v0.4.2 (MVP Stage).
This version includes:
- Fixes for several underlying bugs
- A comprehensive GitHub Account & Gist Setup Guide
- Continued optimization for Gemini and DeepSeek API workflows
Looking Ahead
We are exploring a new path: evidence-based personal branding.
The project is now open-source on GitHub. Because my testing bandwidth is limited, I have primarily optimized and tested the tool with the Gemini and DeepSeek APIs. Since LLM APIs vary significantly across providers, feedback from anyone testing integrations with Claude, OpenAI, or local open-source models would be highly appreciated.
🔗 GitHub Repository: VeriHash