Frannky

8 ms·

121.
by Frannky 8mo ago
I think unless you're doing simple tasks, skills are unreliable. For better reliability, I have the agent trigger APIs that handles the complex logic (and its own LLM calls) internally. Has anyone found a solid strategy for making comp
122.
by Frannky 8mo ago
Have you tried less bloated claws?
123.
by Frannky 8mo ago
I want to use OpenClaw, but it seems like a mess. I want to use glam coding plan as the backend with the since it's cheap. I found ZeroClaw to be an interesting option, maybe hosted on Hetzner. I don't want to give it access to my
124.
by Frannky 8mo ago
I tried Opus 4.6 recently and it’s really good. I had ditched Claude a long time ago for Grok + Gemini + OpenCode with Chinese models. I used Grok/Gemini for planning and core files, and OpenCode for setup, running, deploying, and edit
125.
by Frannky 8mo ago
I feel like they want to be like Apple, and open-code + open-source models are Linux. The thing is, Apple is (for some) way better in user experience and quality. I think they can pull it off only if they keep their distance from the others
126.
by Frannky 8mo ago
I want to learn robotics too!! I have a feeling that trying to build something helpful for myself, with help from LLMs, could be a good strategy—but I have no idea! Possibly budget-friendly
127.
by Frannky 8mo ago
It's interesting how there may be an implicit assumption that imposing more rules on tech will lead to positive outcomes. From my perspective, technology is like reality itself: very difficult to control, with countless ways to achieve
128.
by Frannky 8mo ago
I paid for the $30 plan. It's useful to me via OpenCode as a cheap backend for CLI/Agentic workflows. I also want to try it with Wiggam Loop to test whether they can together build production-level code if guided via prompts and a
129.
by Frannky 8mo ago
I want to give a try to gsd + open code + Cerebras code. Any experience?
130.
by Frannky 9mo ago
Thanks, super useful!
131.
by Frannky 9mo ago
It's free with qwen oauth
132.
by Frannky 9mo ago
I started playing with skills yesterday. I'm not sure if it's just easier for the LLM to call APIs inside the skill — and then move the heavier code behind an endpoint that the agent can call instead. I have a feeling that otherwi
133.
by Frannky 9mo ago
It seems cool! How to use it for free with acceptable quality? Also what are the alternative for a personal assistant that remember stuff automatically and message you about it?
134.
by Frannky 10mo ago
Electronic devices are very effective distraction tools, especially phones. Companies and apps leverage our psychology and biology to get our attention, but we can take control of what we interact with—and if we remove the hooks, they won&#
135.
by Frannky 10mo ago
I see LLMs as searchers with the ability to change the data a little and stay in a valid space. If you think of them like searchers, it becomes automatic to make the search easy (small context, small precise questions), and you won't
136.
Ask HN: Health/Ocean builders: What problem would you pay to get off your plate?
3 points by Frannky 10mo ago|0 comments
137.
by Frannky 10mo ago
Wow! Kudos for thinking it was possible and making it happen. I was wondering how long it would be before big local models were possible under 10k—pretty impressive. Qwen3-235B can do mundane chat, coding, and agentic tasks pretty well.
138.
by Frannky 10mo ago
I am not sure I am doing it the right way or if there's a right way, but what I do is generate small files one at a time, with a functional system of stateless input/output. That way I can focus first on architecture, then on stat
139.
by Frannky 10mo ago
I don't think the author should be sad. He helped push an industry that lets people move value around freely without third parties being able to choke them off. That alone is pretty powerful. The people gambling with it would probably
140.
by Frannky 10mo ago
I left the EU years ago. Today, I feel way happier and freer. Once you're out, you realize how strongly they were influencing your ability to be free via rules, taxes, news, low salaries, and business start-up costs. Try it yourself if
141.
by Frannky 11mo ago
It's a good model. I worry that they will be able to win the game by offering the best service for free, thanks to selling users' data—kind of like search, email, etc. It's sad. Not that the alternatives are better... You eit
142.
by Frannky 11mo ago
I had to stop watching Netflix. Every few minutes, I felt I was taken out of the story by some agenda-pushing content. It's so sad because they are very good at storytelling. I hope they won't enshittify Warner Bros. content too
143.
by Frannky 11mo ago
Pretty smart strategy. I would want a CLI terminal that learns from multiple interactions and updates the playbooks in the CLI loop, so that the agents become smarter after users interactions
144.
by Frannky 11mo ago
The unluckiest MITM attack
145.
by Frannky 11mo ago
Smart model—I use it as my main chat. It's interesting that markets were able to predict that it would lower the revenue of the paid ones.
146.
by Frannky 11mo ago
I haven't tried a Mistral model in ages. Llama and Mistral feel like something I was using in another era. Are they good?
147.
by Frannky 11mo ago
The first models were really bad, but the new models are very good—both text and image. They were able to clean up the embarrassing start and catch up. It's very interesting/scary that a giant corporation can develop things that n
148.
Show HN: Hoodl.net – Find and Vet Top X Influencers in Seconds with PageRank(hoodl.net)
3 points by Frannky 11mo ago|0 comments
149.
by Frannky 11mo ago
I use CLI Terminals as agent frameworks. You have big tech and open source behind them. All the problems are solved with zero work. They take care of new problems. You don't need to remove all the stuff that becomes outdated because th
150.
by Frannky 11mo ago
Cool model. It gets stuff done.

More ›