3 ms·

I’d love to know the specifics on the ppl hiring the consultants and the work output. In my experience you need someone with experience to chain together any ge

I’d love to know the specifics on the ppl hiring the consultants and the work output. In my experience you need someone with experience to chain together any genAI output, as well as validate that it’s even correct. It’s not like it magically does everything for you.

anonzzzies 2y ago
Correct, but that was also the point; all these ‘consultants’ are senior programmers. Nothing is magically done; they use the ai to do the work they were mostly doing before with more people/themselves without ai; now they just focus more on business side. Higher pay and harder to replace (for now). My company does the same but in a niche; our specific tooling is easier to automate (which we did/are still doing) and we notice it is getting easier with the llms getting better. A year ago we still needed a human to evaluate even the basic scope of work, now llama3 + some tooling we built can automatically get us a locally generated report.
- azinman2 2y ago
  > he saw his EU clients cancelling all his contracts because his people could be replaced by openai But this is what I want to know. They thought they could replace the ppl in the first place.
  - anonzzzies 2y ago
    Ah, but those were not consultants or maybe you didn’t mean that. Most of these clients hired people to augment their in house EU teams so they don’t need to hire more programmers locally; after 6-12 months of using copilot/gpt they found out they now could do that work in house with only the in house people in the same time. So they let go of the outsourced team members. I see this happening everywhere. It’s mostly the people that do the ‘light’ work; crud, integrations, frontend, data cleaning that are getting removed; this is a massive portion work wise, but simple, repetitive and boring. Things like business logic, workflows, data science, ml, performance and devops are done in house or still outsourced.

anonzzzies 2y ago

Correct, but that was also the point; all these ‘consultants’ are senior programmers. Nothing is magically done; they use the ai to do the work they were mostly doing before with more people/themselves without ai; now they just focus more on business side. Higher pay and harder to replace (for now). My company does the same but in a niche; our specific tooling is easier to automate (which we did/are still doing) and we notice it is getting easier with the llms getting better. A year ago we still needed a human to evaluate even the basic scope of work, now llama3 + some tooling we built can automatically get us a locally generated report.

azinman2 2y ago
> he saw his EU clients cancelling all his contracts because his people could be replaced by openai But this is what I want to know. They thought they could replace the ppl in the first place.
- anonzzzies 2y ago
  Ah, but those were not consultants or maybe you didn’t mean that. Most of these clients hired people to augment their in house EU teams so they don’t need to hire more programmers locally; after 6-12 months of using copilot/gpt they found out they now could do that work in house with only the in house people in the same time. So they let go of the outsourced team members. I see this happening everywhere. It’s mostly the people that do the ‘light’ work; crud, integrations, frontend, data cleaning that are getting removed; this is a massive portion work wise, but simple, repetitive and boring. Things like business logic, workflows, data science, ml, performance and devops are done in house or still outsourced.