Raphie (@A_raphie) header
Raphie (@A_raphie) profile picture

x rate limited, join the telegram

Oblivion
Joined February 2022
Google when is the reset coming?
2
7
263
Germini 4 Neon suddenly become the best medium sized model was not what I expected at all this year
1
2
181
So that's why the limits is so terrible
Gemini 4 Neon is burning your Antigravity limits ~3x faster. prompt caching is broken (9% hit after 1m) Opus 5.5: ~95% (1h) GPT in Codex: ~97% (24h) sucks cuz Neon is actually insane for agents. Google i know infra is rough rn, but pls fix (put Neon on it lol + reset limits)
282
So never 💀
Project Dawn when?
3
2
22
3,415
Project Dawn when?
2
1
22
6,180
So that's why I'm also suspended from referrals, good to know. 💀
I don’t blame some people who choose to hoard the very opportunities and income streams they benefit from. Let me tell you why. When I posted about Mercor, some of you came into my DMs asking how to get started. I genuinely had no intention of sharing my referral link. Even after making a detailed post explaining how to get started, I still didn’t include my referral link. I wasn’t even particularly interested in the referral program, despite knowing people were earning from it. Then someone commented under my tweet and asked me to share my referral link, saying it wouldn’t make sense to share such an insightful post without giving people my referral link they could use to sign up. So I went to my Mercor profile, copied my referral link, and dropped it under the tweet. Over 12 people signed up through that link. However, when Mercor asked one of the people signing up whether they knew who referred them, this person reportedly selected “No” and provided other information. This morning, I discovered that I’ve been banned from the referral program. Honestly, I’m okay without the referral program. I was never depending on it. But this is exactly why some people prefer to keep certain opportunities to themselves. You try to share something helpful, and someone turns around to sabotage you. If you’re the person who reported my account, I hope you know that Mercor won’t give you shishi for it. That’s all I have to say for now.
2
7
1,458
Lmao crazy day that day Woke up with 2 contracts and slept with 0 💀
Mercor fairs man ggs
15
1
93
7,201
Yes it does
People who fuck butthole, does the dick come out smelling like poop?😭😭😭
5
4
1,231
Mercor fairs man ggs
5
1
24
8,885
I need a good/cheap uncensored model man
5
1
11
1,113
Raphie retweeted
New opportunity on Mercor! If you previously worked as a reviewer on Project O, here's another opportunity worth checking out. There are two exercises to complete: 1. Rubric Review Exercise 2. Task Review Exercise What does the role involve? You'll be reviewing AI-generated tasks, including prompts, grading rubrics, and model responses, to determine whether the model's failures are genuinely justified. Your responsibilities will include: - Verifying that the AI agent had access to all the information needed to complete the task. - Checking whether grading rubrics accurately reflect the user's instructions. - Identifying situations where a valid response could be unfairly marked as incorrect. - Ensuring rubrics comply with the required evaluation guidelines. - Rewriting unclear, unfair, or poorly structured criteria. - Flagging tasks that cannot be fixed and briefly explaining your reasoning. The goal: Ensure AI-generated tasks are evaluated fairly, consistently, and strictly according to the actual instructions. If you have experience reviewing AI outputs, this is definitely worth exploring! t.mercor.com/sOlQF
4
3
1
30
2,382
New opportunity on Mercor! If you previously worked as a reviewer on Project O, here's another opportunity worth checking out. There are two exercises to complete: 1. Rubric Review Exercise 2. Task Review Exercise What does the role involve? You'll be reviewing AI-generated tasks, including prompts, grading rubrics, and model responses, to determine whether the model's failures are genuinely justified. Your responsibilities will include: - Verifying that the AI agent had access to all the information needed to complete the task. - Checking whether grading rubrics accurately reflect the user's instructions. - Identifying situations where a valid response could be unfairly marked as incorrect. - Ensuring rubrics comply with the required evaluation guidelines. - Rewriting unclear, unfair, or poorly structured criteria. - Flagging tasks that cannot be fixed and briefly explaining your reasoning. The goal: Ensure AI-generated tasks are evaluated fairly, consistently, and strictly according to the actual instructions. If you have experience reviewing AI outputs, this is definitely worth exploring! t.mercor.com/sOlQF
4
3
1
30
2,382
Raphie retweeted
I have vouched for every account I referred on Mercor. I hope that would help them get an offer
17
1
63
2,924
Openai sending more emails than released models us wild
1
4
436
I have vouched for every account I referred on Mercor. I hope that would help them get an offer
17
1
63
2,924
Check your DM
you asked for a referral program, so we built one! 🎉 earn $20 for every new Guild user who signs up through your link, qualifies, and completes their first approved, billable hour. log in to Guild talents.getblend.com/ → Affiliate Program to get your unique referral link. we're launching in beta, starting with AI Red-Team Specialists only. share with friends, colleagues, or your community. a few rules: new Guild users only registration must happen through your link one referrer per person no self-referrals or duplicate accounts bonuses are credited once the qualifying work is approved let's see what we can build together!
332
Raphie retweeted
David Hudenyin what a guy
4
1
5
587
David Hudenyin what a guy
4
1
5
587
My account is a mess man. God 💀
1
1
535
Raphie retweeted
What a way to start the day Bro cooked ngl 😭
If you've worked as a generalist in any Mercor evaluation project, and are an active user of Gemini or Instinct, Hit me up with your expert email for direct placement
120
7
1
465
41,417