A coding monkey (Software Engineer) exploring the Great White North

Joined October 2014
66 Photos and videos
Pinned Tweet
"A lot of people are very uncomfortable with the fact that they are essentially just CRUD monkeys. They just make systems that create, read, update or delete rows in a database, and they have to compensate for that existential dread by overcomplicating things." - @dhh
48
57
811
237,111
Everyone I know personally who’s doing really good in their career has some passion outside of work.
41
Everyone enjoying start of summer 🌞
34
just noticed that whole day had at least one agent doing something claude (for learning) -> copilot (work) -> hermes (personal stuff)
1
1
115
so I created a personal AI Assistant with @NousResearch hermes agent, running on most basic droplet by @digitalocean and my @GitHubCopilot subscription Using the agent since last 2 days without any cost. @digitalocean gives you $200 credits on new account
But can the personal AI agents add actual value? I can try to create an AI assistant and then replace all my productivity, habits tracker apps with the AI assistant Later I can build graphs to visually see how am i progressing, this can be cool side project.
1
142
no premium requests so far
50
But can the personal AI agents add actual value? I can try to create an AI assistant and then replace all my productivity, habits tracker apps with the AI assistant Later I can build graphs to visually see how am i progressing, this can be cool side project.
191
The personal AI agents are getting so cool, especially Hermes Agent by @NousResearch, I played around with a bit on my local machine and it was really easy to setup. I created a docker image with all the tools required to install hermes agent quickly, if anyone want to give it a quick run
3
100
Replying to @NousResearch
Here is the docker image: hub.docker.com/repository/do… It acts like a VPS, so if you later want to move your agent from local to actual VPS, you can do it easily

69
Honestly, building AI Apps is genuinely hard, not a demo, but actual production-level software. Like, how do you even test the behavior of a probabilistic system? and make sure it's reliable Even more difficult is building AI Apps that provides actual business value.
1
1
42
Good place to start with neural networks
26
Now that I think of it, I don't really need IDE anymore, I need something really light weight and fast which just shows me code and diffs of change that agent did. Maybe I should try vim again.
29
Shaik Saifuddin retweeted
OVERRATED: running tons of agents in parallel; working on too many things at once; perpetual context-switching; opening lots of low-quality PRs that may never land. UNDERRATED: using one or two agents at a time; focusing on the task in front of you; thinking deeply; finishing stuff; making your code works in prod.
220
393
4,979
247,017
ok, solving DSA using python helps in learning python properly. Coming from Java, I love how clean it is — no massive syntax, no unnecessary imports everywhere. But one thing felt so weird to me… why do booleans in Python start with capital letters? True and False just look off 🤔
I always avoided python, it's syntax was never intuitive for me, and almost everything that can be done in python, I was able to find workaround with JS/TS, but looks like there is no running away from python now. Maybe I should start practicing DS/Algo in python, Idk why I still do that in Java
57