Badge gives an AI agent that collects peer reviews to build proof of work. For job seekers - The agent connects to your contacts, requests anonymous reviews from your past colleagues, and builds a verified, portable trust score. score. For hiring managers - Badge gives you a 30 second reference checks on candidates, collecting authentic feedback from their actual coworkers to replace unreliable, AI-generated resumes and LinkedIn Recommendations.
Forum Threads
View all
DiffSense
background checks / references are notoriously unreliable. Stale, outdated, fake, shallow etc. So its definitly a pain to try and solve. I think maybe connecting it more to linkedin could be useful. Like a review from a strong linkedin profile goes a lot longer than 10 goood reviews from noob profiles etc.
Badge
@conduit_design Absolutely, and thanks for pointing this out. So our trustscore works the same way. When a user with higher trustscore gives a review for someone, the trustscore of the reviewee will have a higher boost. Do you have any feedback requests?
DiffSense
@lokesh_motwani1 What is the trust-score made up of?
Badge
@lokesh_motwani1 @conduit_design The Trust Score is calculated using authenticated review from colleagues focusing on essential traits: work quality, reliability, professionalism, communication skills, and resilience under pressure. Key influencing factors - Volume: Accuracy increases as you collect more reviews, Recency: New reviews are given higher priority, Reviewer Credibility: Reviews from users with high trustscore carries more significance andConsistency: Recurrent themes identified across different reviews are heavily weighted.
New users begin with a baseline score of 60. As you maintain a presence on Badge, your reputation grows and compounds, making your professional standing increasingly credible over time.
DiffSense
@lokesh_motwani1 @raaghav_naraayan_m_v Is that system relible tho? It seems easily gameable. Imo score only works when it has scarcity built in. Any system that isnt tied to scarcity will be gamed and thus loss its value. Very hard to solve of course. Im sure you have some ideas on how to solve this in the future. Will be interesting to follow and see how it develops in the future!
@raaghav_naraayan_m_v pairing anonymous submission with work-email verification is smart to fix the honesty problem LinkedIn recommendations have. but how does that work for freelancers and consultants who move between short-term clients rather than a single employer? curious whether you can build a meaningful Trust Score without one fixed email
Badge
@clement_avq Great question! So for that the clients can always drop a review using the linkedin profile of the freelancer they worked with. Apart from that we are also planning to partner up with freelance platforms such as Fiverr,Upwork to bring in more credibility.
@raaghav_naraayan_m_v Great, thanks for the feedback!
The timing for this is right. I have watched the resume lose its signal value this year (200+ applications per role, most AI-polished, all reading identical), and the honest answer to what replaces it has always been some form of verifiable work. Peer reviews collected by agents is an interesting take on that. One question: how do you keep the peer reviews themselves from becoming the next gameable artifact, once people know an agent is collecting them?
Badge
@virko_kask Great question Virko, You are right, Bad users will try to game the system and we have to stay one step ahead always. We will use ML in calculation of trust scores
@lokesh_motwani1 Thanks Lokesh. One thing I would gently push on, from building scoring systems myself: when the trust score comes from ML, the gaming problem moves inside the model, and users can no longer tell a fair score from a gamed one. Whatever the model computes, show the receipts behind it. People act on scores they can verify and quietly ignore the ones they cannot. Rooting for you, the proof-of-work direction is right.
You're right when it comes to being anonymous, you're more than being honest but at the same time it could also remove accountability so how do you prevent an unfair review from ruining a person's reputation?
Badge
@reda_roqai_chaoui The user does get the option to hide the review if it's truely unfair with that being said there is a limit to how many the user can hide as well.
Documentation.AI
Unique product! How does Badge handlesfreelancers or consultants who work with many short-term clients instead of one employer?
Badge
@roopreddy It still does work the same way wherein the freelancer/consultants can ask for a review from the client, therefore building the trust score just like any other regular candidate.
Badge
@roopreddy Thank you @Roop. Badge is a portable review system for freelancers. Clients can give review to the freelancers using LinkedIN, email or phone number.
Badge
The hard part here is making the proof portable without turning it into another opaque score. Peer context, recency, and the option to inspect what evidence produced the badge would matter a lot for trust, especially as AI makes resumes easier to polish.
Badge
@krekeltronics absolutely, we did consider that. It all boils down to how the trust score is calculated.
Your Trust Score is built from verified peer reviews across key dimensions like: quality of work, reliability, professionalism, communication, and how you handle pressure.
What affects it:
Volume : more reviews = more accurate score
Recency : recent reviews carry more weight
Reviewer credibility : reviewers with stronger Badge history carry more weight
Consistency : patterns across multiple reviews are weighted heavily
Continuous peer review as proof-of-work is a clever inversion of the reference letter. What stops it drifting into LinkedIn-endorsement inflation where everyone five-stars each other? Genuinely curious how the incentive design handles that.
Badge
@chielephant Thats a wonderful point. Incentive design does not handle this but anonymity of reviews handles this. For linkedin and uber reviews. The feedback giver does not want to be in bad relation with the reviewee because it is public. When we make all reviews anonymous then there is no politeness and people can be honest and thats why we feel, it should not drift to 5 star clusters.