Georgios KonstantopoulosX (Twitter) profile trust report
very greek accent. general partner & cto @paradigm. eng lead @tempo. optimist and rustacean š¦ // team lead of reth, foundry, alloy. join us.
Algorithmic trust score
Looks genuine
Higher scores indicate fewer risk signals.
- Last assessed
- Score based on
- 1 observer
- Engine version
- vv1.6.0
- Evidence coverage
- Limited signals
Some signals were unavailable. Read this score with extra caution.
How is this score calculated?Public signals on this account are consistent with a genuine person or organisation.
This report uses public signals. It cannot verify identity or guarantee safety.
Your account? Request a review or removal
Scan activity
- Scanned by
- 1
- xlegit user has shared evidence for this profile.
- Account age
- 9 years
- On X since November 2016.
- Published assessments
- 1
- First scored 21 Sept 2026.
- Community reports
- 0
- User reports, shown separately from the algorithmic score.
Topics and tone
This sample often discusses technology & software. A consistent tone is not clear yet.
Recurring topics
Technology & software
Writing tone
Not established from this sample
Formats
Stories & reflections
Based on 20 sampled own posts. This describes the writing, not the person or their expertise. It does not affect the trust score.
Read the sample
here's the only frame of the teaser you need to see to know that you must apply to frontiers and come hack in person with us - it's free, it's open source, and it's all deep engineering w/o sponsors or booths. https://t.co/2mXvbBIiCR https://t.co/Gz3Lqhre0d
every paradigm event has had PSA graded-style casings for unique metal cards for event attendees. i've been collecting them, and it feels really cool once you have a few of them. https://t.co/Gz3Lqhre0d
https://t.co/Z9mmMeFCxD
frontiers merch is gonna be sick applications close this week 1 more very special speaker to be announced fort mason sf oct 12-14 https://t.co/yfSwpR7V1H
continuing cooking on the auto-router w/ jev, my latest prompt after integrating all gateways "make Jev auto-route across all, taking latency into account w/ TTFT test probe prompts that run every some interval" https://t.co/ZB2rIof3Q9
@grok @tobi lmk if u wanna chat @tobi, i think it's starting to cook into something really strong https://t.co/ara0tG6tZS
@tobi @grok explain what gakonst/nanocodex does about the last paragraph w/ our just-bash stuff incl how all the connectors work
@0xInuarashi @mediaquery companion app that writes the ai chatlog to a file?
@rakyll @zhygis
@ETazou cool! i was thinking of infinity blade so this tracks hehe
@meanjanroy not same wifi
experimenting with jev-backed model routing, opening prompt gets mapped to an eval, then choose which model has best cost / completion time for eval, can auto decide or can expose to user the cost latency tradeoff https://t.co/pFJl11MTCP
@jconorgrogan how are we gonna audit the spec? i think the issue is i myself cant lay out 100% of the behaviors/edge cases in our software- not sure most our team can if you can define the system you will be able to provably codegen optimal and correct code- but idt ppl can define things
@GiulioRebuffo word ill get that running rn and lyk where we are tmr
@VictorTaelin check this out! i think the crypto world rly gets why this stuff is impt https://t.co/vfPpSWBVw7
my bend evm seems to pass all the Amsterdam Ethereum HF state tests - its extemely slow but its worth taking a look as i diff tested w revm/evm2 and looks good https://t.co/7J8GPmfyLY
@juanbenet š
Omarchy desktop over WebRTC w a Swift app on my phone - sub 100ms delay, need to polish the controller UX but genuinely astonished, you can build anything https://t.co/5ki6fGh6kY
@joshxhuang yeah super real- webrtc powered
@thorstenball makes total sense
Open this profile in the extension and choose āAdapt this style for my accountā to build a voice guide and content plan from your own ideas.
Before you trust, follow or reply
A closer look at what the available evidence tells us, and what it leaves open.
Still an open question
Scams & suspicious requests
The available checks did not surface a clear scam warning. That does not tell us whether every offer or link is safe.
Check the destination and the terms of any request for money or personal information.
Still an open question
Identity & impersonation
We have not established who runs this account. A familiar photo, a badge or a high score cannot settle that on its own.
Use the personās or organisationās own website to confirm the account when identity matters.
Still an open question
Useful, original content
There is not enough evidence here to say how much original, useful detail this profile adds.
Read a few complete posts. Look for clear examples, sources and answers that help with your interests.
Still an open question
Rage bait & clickbait
The available checks did not establish a repeated bait pattern. A short sample can still miss it.
Look for a clear point or useful detail behind a dramatic headline.
Fig. 01
The signals behind the score
Account age
The account has an established history. That gives you more context to review, but older accounts can still change hands.
Follower / following ratio
This part of the available activity adds reassuring context. Read it alongside the other checks; it is not a guarantee of safety.
Mix of originals, replies and retweets
This part of the available activity adds reassuring context. Read it alongside the other checks; it is not a guarantee of safety.
Fig. 02
Score breakdown
Each bar shows the points earned in one dimension. Missing dimensions were not assessed. A hard risk signal can override the combined result. Read about the signals on the algorithm page.
- Account fingerprint25 / 25 pts
- Content originality19.9 / 25 pts
- Behavioral patterns13.7 / 15 pts
- Network quality12.6 / 15 pts
- Engagement authenticity9.0 / 15 pts
- Visual identity3 / 5 pts
Fig. 03
Score history
One daily snapshot so far. This chart records the published assessment for each day with a score.
Show as a table
| Date | Score | Observers |
|---|---|---|
| 21 Sept 2026 | 84 Ā· Looks genuine | 1 |
Fig. 04
Community reports
These reports come from xlegit users. They appear alongside the trust score and do not change the number.
- Active reports
- 0
- Distinct reporters
- 0
- Community score
- 0/ 100
- A higher score means stronger concern from community reports. Each report is weighted by the reporterās history.
There are no active community reports for this profile. Users submit reports from the extension. No reports does not guarantee that an account is safe.
Reading the report
What this report can tell you
Can this tell me whether @gakonst is a bot?
It can help you spot signs of automation, fake accounts or scams, but it cannot prove who is behind an account. A high score does not verify identity. A low score does not prove wrongdoing. Read the reasons and check important claims for yourself.
How current is this assessment?
The published score was assessed on . This page refreshes hourly. Reloading it does not run a new scan. To request a fresh check, open the profile in X with the Chrome extension installed.
Why might community reports disagree?
Reports reflect usersā experiences and are weighted by their reporting history. They are separate from the trust score and have not been independently verified. If no reports appear, it means no active reports are available in xlegit. It does not mean the account is safe.
Check profiles while you browse X
Check profiles while browsing X with the xlegit Chrome extension. Free includes 5 scans a day, up to 30 a month, plus 15 post checks a day. Pro adds 200 daily post checks, Link guard, Focus mode and timeline badges.
Is this your account?
If you think the score is wrong, you can ask us to review it. You can also ask for your score to be withheld. We will explain our decision and correct the assessment when needed.
Request a review of @gakonstReport generated 21 Sept 2026, refreshed hourly. Not affiliated with X Corp.