Deependra Solanky (@solanky) header
Deependra Solanky (@solanky) profile picture

Product. Tech. Open Source. Life.

šŸ‡®šŸ‡³
Joined February 2008
Photo by Deependra Solanky (@solanky)Photo by Deependra Solanky (@solanky)Photo by Deependra Solanky (@solanky)Photo by Deependra Solanky (@solanky)Photo by Deependra Solanky (@solanky)Photo by Deependra Solanky (@solanky)Photo by Deependra Solanky (@solanky)Photo by Deependra Solanky (@solanky)Photo by Deependra Solanky (@solanky)Photo by Deependra Solanky (@solanky)
Deependra Solanky retweeted
Some of you have been asking very politely. Some of you have just been asking. GPT-6.2 Astra and Sol are now available. Super efficient and built to perform where it counts—the real world. openai.com/index/introducing…
1,523
2,314
2,203
23,496
2,776,644
Deependra Solanky retweeted
Meet Devin Bots. From writing firmware to merging thousands of PRs a week, a Devin Bot leads a team of agents to get the job done.
205
217
287
2,061
321,819
After Outlook emails not working since last few hours now Claude Code is also not working.
659
Sometimes feature removal is better than adding features. Dial was confusing.
Bye bye, Dial. low, medium, high, and ultra are retired. In their place is a model selector: pick your model, switch mid-thread, and see exactly who gets billed. ampcode.com/news/bye-bye-dia…
11
Deependra Solanky retweeted
Which Claude model should you build with? Our new guide breaks down when to use Fable 5.2, Opus 5.5, Sonnet or Haiku. It also covers when to adjust effort instead of switching models, and how to save cost by putting subagents on smaller models. claude.dev/blog/which-claude…
371
1,224
304
16,217
2,594,418
Deependra Solanky retweeted
Introducing Claude Fable 5.2 and Claude Mythos 5.2, designed to tackle the most daunting technical challenges. They’re capable collaborators on ambitious projects in many fields, and show improvements in areas like coding, reasoning, and vision.
3,175
9,246
6,961
96,277
20,230,880
Claude Projects' coordinator implementation feels quite inferior to Amp Spaces. Claude's coordinator creates a separate thread for each task, with little meaningful coordination afterward. It ends up being just a postman passing tasks around, defeating much of the purpose of having a coordinator. In Amp Spaces, threads communicate much better with the coordinator, making the whole experience far more cohesive. After using both, Amp's approach feels much closer to what multi-agent coordination should be.
1
1
39
With Conductor announcing its shutdown today, Amp's 3-month free subscription offer comes at an interesting time. I never tried Conductor since it was Mac-only, but I've been meaning to give Amp a proper try. Just redeemed the offer. Looking forward to testing Amp on real coding work and seeing how it compares with my current tools.
heard a lot of people are leaving conductor. Good news is that amp is RIPPING and we'd love to have you give it a go, i'll keep this short: 1. download the mac and ios app ampcode.com/app 2. three months of free sub below 3. import all your threads from all your coding agents
50
One thing I really miss after switching from Codex Desktop to Claude Desktop last month is automations. Claude’s scheduled jobs are nowhere near as good as Codex’s. Most of my Codex automations were Heartbeats, and Claude doesn’t have a real equivalent. I’m now experimenting with Hermes bots and cron jobs to recreate that experience. The added benefit is portability. I don’t want my automations tied to whichever coding agent I’m using this month. If I switch between Claude, Codex or something else, they should keep running without needing to be rebuilt.
3
2
264
One frustrating limitation of Claude Code Remote Control on mobile: you can’t create new sessions from a coordinator session. You have to go back to your computer to start them. It defeats part of the purpose of remote control. If I’m coordinating agents from my phone, I should be able to spin up new sessions too.
3
1
49
Uninstalled Orca today and moved fully to the new Herdr desktop app. Orca served me well and the team ships fast, so no complaints there. What tipped it for me was how each tool tracks agents. Orca puts hooks into every agent's config. When I cleaned up, I found 13 in Claude Code, 9 in Codex, 8 in Devin and a plugin in Hermes, all running on every prompt and tool call. Herdr reads agent status from the terminal itself. My Claude and Codex configs are clean again: no hooks, nothing running between me and the agent. What I like most: • Agents can drive the workspace through a small CLI. Claude opens a pane next to itself and leaves a command at my prompt (sudo installs, prod releases) without running it. I just press Enter. • Spaces, tabs and panes per project, with Claude and Codex side by side • Less setup and fewer moving parts I still use the Claude and Codex desktop apps too, but for terminal agent work Herdr is home now.
40
4
1
240
25,813
This was much needed on Linux. Codex and Claude lack computer use on Linux.
Introducing agent-desktop agent-browser, for your whole computer → Computer Use for any coding agent → macOS, Windows and Linux → Runs in bg while you keep working → Agents use your computer in parallel, each with its own cursor
88
I am verified and 5-hour limit is removed :)
48
Herdr desktop app launched
*ehm ehm* BIG ANNOUNCEMENTS DAY šŸ“£ two new things cooking in herdr labs šŸšŸ§Ŗ a brand new herdr-core, and a desktop app built on it. open source. first, the app: software that fits you. every part of it is a file your agent can change. you just ask.
1
159
Deependra Solanky retweeted
1/ Excited to announce our new company Particle This is our first step in a series of models focused on developers. With particle-1, we focused on token efficiency and managed to keep parity in performance while using ~45% fewer tokens. Live now: particle.ai
272
128
117
2,260
439,153
Testing GPT-6.1 Terra on the Plus plan. Not sure about token usage compared to Sol yet, but overall it's very fast and browser use feels super snappy. Haven't tested it on coding since I already have Claude Max for that.
59
Deependra Solanky retweeted
GPT-6.1 Terra: delivering fast and capable computer use at a lower price. GPT-6.1 Terra understands what’s on screen, takes action across apps, writes naturally, and is a more capable model for your whole workday with the efficiency to take on more of it.
šŸ¤– Made with AI
2,241
2,560
2,339
28,852
8,331,829
Stopped using Claude Code Projects and moved back to a Chief of Staff workflow on my local machine. Claude Projects asks for too many permissions, interrupts often, and switching back and forth between threads is cumbersome. Hope these get fixed once it's out of beta.
1
1
65
Now I have to decide whether to keep Codex by downgrading to the $20 Plus plan or continue with Devin at $20. Devin is good, but I haven't even used the free SWE-2 in days. Its real appeal for me is Cloud, where $20 isn't enough. For CLI/Desktop, Claude already covers my needs. Codex seems more useful as a backup, especially for reviewing Claude's work.
I still have a week left on my Codex subscription, with 3 banked resets, and Devin SWE-2 is free too. But since switching to Claude, I haven’t felt the need to use either. Claude’s usage limits have been more than enough for my work so far.
82
Just got access to Claude Code Projects. Watching a coordinator thread orchestrate parallel threads in real time is a massive workflow upgrade. A few rough edges here and there, but overall a seriously useful addition.
40