I found a link while searching Google for this company that was offering $50/hr for a generalist position doing what I figured was standard RLHF (Reinforced Learning from Human Feedback) annotating. My first thought was "yeah Scammmy Mc Scam Scam over here gonna pay $50/hr to annotate some data, suuuure." That is because I have been involved with a couple other companies that do not like to hold true to their promises and have no ethics when it comes to contractors rights. Not salty, just very very sodiumy.

I decided to check into the company because I had seen some posts on reddit for people getting contracts for 100-150/hr with mercor ai I believe it was. Turns out the company is legit and  have their own niche style of training and annotating. I saw a couple articles refer to system as "AI training Gyms". Immediately the Pokemon theme song started playing from some diegetic source that I still cannot pinpoint. So I applied, took the assessment and passed.

I was pretty stoked at this point because I normally do terrible at these types of assessments due to the fact that I refuse to read the guidelines until I have already failed. However, this time I was stricken with curiosity before I even applied. Their process is legit and completely different  than any other annotating I have done in the past. The position is best described as a prompt creator, but that isn't the entirety of the job specifications. I cant say too much, but you not only create the prompt but you have to act it out as well. 

The guidelines for the tasks are very technical and require a good amount of creativity to complete. As I was saying, I passed the assessment and within a day or so I was invited to complete the tryout tasks. Three tasks that are exactly the kind of tasks that you would be doing in production. Out of the three I had one accepted my first try and two had to be fixed due to the prompts being ambiguous. Which basically means that there was more than one possible outcome. The reviewers are all very understanding and very professional from my experience so far. They tell you exactly where your prompt needs work, why it did not meet expectations, guidance on how to improve it and are very encouraging. I fixed the two prompts and they were accepted a few hours later. 

A day or two later I was invited to the onboarding and sent a link to sign a contract for $50/hr as a generalist. That was 11th, and since then I have completed one pay cycle which will be paid out through Deel.com on the 22nd and for only five days worth of work I earned 1461.68 for the pay period and today I put in 5 hours and have $300 towards the next check already. 

There is no guarantee, as anyone who has been in this field knows that work will stay steady. However, for only being here for a week and having a very very sodiumy bias towards AI training or data annotation companies, I see a monumental potential in this company and the way carry themselves. That is before even noting the their state of the art process for RLHF. If anyone thinks they would be a good fit I will leave my referral link for you to apply.

Jump on this opportunity while it is here. Regardless of what I think, you never know how long these jobs will be around for.

https://fleetai.com/start?referralCode=c5d3f24eef35480785d2b6a148d5224fcinhwo