mlcruz
I also do believe search quality is recall dominated, but the link between the metrics and results are not as clear as they could have been. I think you have a stronger sell here than it looks at first glance.
Congratulations to the team tho! Those are very nice results
Can we keep at least this place clean of this kind of discussion?
Both yours and the other guys side are having this same discussion on every single platform possible. Do you guys really need to do it here?
I remember that close to an year ago i felt the same way about opus 4.5. And even if you can easily run ds or qwen locally to get very similar results to that, we moved to 5.5 being the new target.
Deepseek can do 90% of the job but a claude reviwer/spec writer is a brutal improvement.
Porto Alegre metropolitan area is having a huge outbreak. My girlfriend is a vet and has been dealing with new cases multiple times a week.
Many of our friends also got it (it is very hard to not get scratched when handling with cats in pain).
It is a really really shitty, painful and hard to treat disease, requiring multiple months of treatment. It is very painful but usually not letal for humans and cats that are in the earlier stages and get treatment.
However it is absolutely lethal for populations of wild and stray cats, as it is very infectious and 100% lethal unassisted.
>Reach for a workflow when a task needs more agents than one conversation can coordinate, or when you want the orchestration codified as a script you can read and rerun. Examples include a codebase-wide bug sweep, a 500-file migration, a research question that needs sources cross-checked against each other, and a hard plan worth drafting from several independent angles before you commit to one.
https://code.claude.com/docs/en/workflows
The results are good, but it is very expensive. I used a workflow to do a full review of my entire codebase, it spawned 75 agents and surfaced and fixed some (real) bugs. It feels a bit overkill, but it works.
Sadly since fable usually works comfortably for 10-20min at time without human input, i end up juggling at least 3 other agents and it lasts me about 2 hours.
If i have a really hard problem or big refactor, i use workflows. This consumes the entire session quota in about 45 minutes.
Keep up the good work!
When im in implementation sessions i try to not let the llm do any decision making at all, just faster writing. This is way better than manually typing and my crippling RSI has been slowly getting better with the use of voice tools and so on.
Yeah i misunderstood your point, i completely agree with what you are saying.
I honestly do not believe that strategy, decision making and other real life context dependent are going to be replaceable soon (and if it does, its something other than llms).
> I'm not sure that I agree with white collar jobs being done for, not every process has as little consequence to getting it wrong as (most) software does.
Maybe im too biased due to working in a particularly inefficient domain, but you would be surprised how much work can be automated in your average back office.
Much of the operational work is following set process and anything out of that is going to up the governance chain for approval from some decision maker.
LLM based solutions actually makes less errors than humans and adhere to the process better in many scenarios, requiring just an ok/deny from some human supervisor.
By delegating just the decision process to the operator, you need way less actual humans doing the job. Since operations workload is usually a function of other areas, efficiency gains result in layoffs.
Trust me, this is a work in progress. Right now most corporations do not have their data organized and structured well enough for this to be possible, but there is a lot of heat and money in this space.
Imo, What most of the people that are not directly working in this space get wrong is assuming swes are going to be hit the hardest: There are some efficiency gains to be won here, but a full replace is not viable outside of AGI scenarios. I would actually bet on a demand increase (even if the job might change fundamentally). Custom domain made software is cheaper as it has ever been and there is a gigantic untapped market here.
Low complexity to medium complexity white colar jobs are done for in the next decade through. This is what is happening right now in finance: if models stopped improving now, the technology at this point is already good enough to lower operational costs to the point where some part of the workforce is redundant.
Your article is great! As someone who's working in this space, your points just improved our presentation and selling a lot. We have been talking with C level finance executives about building semantic layers, and i can confidently say that the way you presented the value proposition of the context layer is going to improve our conversion rates.
Thank you so much! This is one of the best analysis i have ever heard about the subject.
Right now we handle all of our recordings internally, since having a bot join our calls is a dealbreaker, but it is something that i'm going to take a closer look at. Your product is very interesting for our problem domain, but having a bot join our calls is a big no.
About the compliance recording:
Microsoft teams allows for two modes of call recording (https://learn.microsoft.com/en-us/microsoftteams/teams-recor...):
- Convenience recording- an ad-hoc recording of a call or meeting that a user starts and manages. For an overview of convenience recording, see Overview- Recording and transcription for Teams meetings and calls. -> This requires a bot to join the call
- Compliance recording-calls and meetings that are automatically recorded without user intervention and owned by the company, using a third-party solution. -> This does not require a bot to join the call. The audio is streamed directly to the server, no bot required.
It is somewhat painful to roll your own compliance recording bot without a third-party, but it is possible (even if badly documented).
I see that one of the pains points of your product is related to having an notetaker bot joining the call. You could probably offer to your enterprise customers that use teams a better experience by processing calls via compliance recording policy (since it does not require a bot to join the call and the user approval flow is way less disruptive).
If it is something that interests you we could talk more in detail about how the implementation works and the its challanges
Rio de Janeiro is by far the most corrupt Brazilian state. Its hard to explain how bad it is if you are not Brazilian, but imagine that every single former state governor and many of the mayors have been sent to prison for corruption after their term ended.
So what usually happens is that someone from the public sector opens up a public bidding for some service to be done by the private sector, and usually who wins is someone who has ties with the local government.
Most of the time whoever wins the bid (usually some shell company) is going to barely offer the service, and share most of the profits with their associates in the local gov.
This is one of such cases: The private lab doing the tests is owned by the cousin of the former state secretary of health Dr.Luizinho. Its very likely that they just did not do the tests at all (yes, that how bad it is)
Just another normal day in Rio de Janeiro.
I would guess that close to 10% of the codebase at my company is straight up copy pasted from ChatGPT without further thought.
So the second person knows that the first person is ready
https://en.m.wikipedia.org/wiki/List_of_countries_by_GDP_(no...
By far the best place to live in Brasil is São paulo (state) countryside. Very safe cities, close to urban centers, not expensive at all. Santos is also pretty nice but 1k might fall short.
https://en.wikipedia.org/wiki/List_of_municipalities_in_S%C3...
Pretty much every single firefox user i know hates proton, every single update feels very hostile to 'power' users. Its also terrible for people with bad eyesight (No tab separators, bad contrast).
Its is impossible that Mozilla is not aware of how hated proton is overall. I really don't want to use anything chromium based, but at this point they are almost forcing me.
Nomad/Consul/Vault/Fabio are super easy to setup, and its like 10% of the effort of handling your own kubernetes.
Documentation could be worked on, but its getting better.
Local ISPs are great for competition.