All activity
shemith mohananleft a comment
Interesting combo of AI app generation + built-in distribution. The “describe → ship → discover” loop feels powerful if the output is truly production-ready. Curious how much control builders have post-generation (code ownership, edits, infra) as projects grow.

SpawnedBuild, launch, and discover products in a single platform
shemith mohananleft a comment
This is interesting because it shifts the problem from “better prompts” to better outputs. Generative UI feels like the missing layer between LLMs and real products — text is rarely the final interface users want. Curious how you’re handling state + re-renders when responses evolve over time.

Agent Builder by ThesysBuild AI agents that respond with UI instead of text
shemith mohananleft a comment
Love the positioning here — “own the stack, don’t rent it” will resonate hard with builders burned by SaaS sprawl. The focus on local AI + self-hosting + real configs (not fluff) makes this feel practical, not aspirational. Curious how members are actually using Clawdbot day-to-day.

Tinkerer ClubThe private club for ppl who automate, self-host, and use AI
shemith mohananleft a comment
This is a great take on scheduling. Email is still how most real scheduling happens, and automating that instead of forcing everyone into links feels much more natural. If Ting handles messy threads well, this could remove a lot of back-and-forth friction.

Meet-TingAI that gives your schedule a brain.
shemith mohananleft a comment
Impressive direction. The real value here isn’t just “bigger model,” but how naturally Gemini works across modalities. If the text–image–code handoff feels truly seamless in real-world workflows, this could change how people actually use AI day to day — not just experiment with it.

Agentic Vision in GeminiAgentic visual reasoning with code execution
shemith mohananleft a comment
This is super useful — being able to ask questions with actual source-backed answers solves a real pain. Curious how you handle large repos and context limits?

ForumsAI-powered Q&A for GitHub repositories.
shemith mohananleft a comment
This is such a delightful take on focus tools 😄 The evolving mascot + streaks combo is really smart — it adds just enough emotional reward to keep people coming back. Love the idea of “growing focus” instead of forcing productivity. Nicely done!

Mochi FocusYour focus timer that grows with you
shemith mohananleft a comment
This is interesting — especially the “iterate until it feels right” angle. Thumbnail design is one of those things where speed + feedback matters more than perfection. Curious how you’re handling style consistency across a channel (or if that’s on the roadmap). Nice execution 👏

Thumbfa.stMidjourney for YouTube Thumbnails—your face, every time
shemith mohananleft a comment
Nice to see the Gatsby team tackling AI workflows. The combo of memory, evals, and tracing in one framework feels very timely. Curious how Mastra compares to rolling this stuff manually.

Mastra 1.0Build AI agents with a modern TypeScript stack
shemith mohananleft a comment
That’s a bold claim — but honestly, it feels like the direction things are going. Curious how much customization you get after the initial generation.

FimoWhere your website lives, collaborates, and evolves with AI.
shemith mohananleft a comment
Turning intent into action is the real bar for AI agents. If Manus can reliably execute end-to-end tasks (not just suggest steps), that’s a meaningful shift. Interested to see how it handles real-world edge cases 👍

Manus Meeting MinutesFrom in-person meeting to finished work in one flow
shemith mohananleft a comment
Research discovery is painfully noisy—this feels like a solid step toward signal over volume. Curious to try the context-aware ranking.

Chirpz AgentSmartest way to discover unseen literature
shemith mohananleft a comment
Clean idea. Search is something everyone uses, but few rethink properly. Interested to see how this improves relevance and discovery in practice.

Gmail in the Gemini EraAsk your inbox anything to get summaries, drafts, & more
shemith mohananleft a comment
Natural language as the interface is the right direction. Removing setup complexity is what will unlock real usage for non-technical teams. Curious to see how Instruct handles reliability and edge cases as workflows scale.
Instruct 2.5Connect apps, describe the job, then automate your work
shemith mohananleft a comment
Impressive direction. On-device speed + efficiency is where real adoption happens, especially for privacy-sensitive and latency-critical use cases. The hybrid architecture angle is interesting — curious to see how LFM2 performs in real-world edge scenarios compared to current lightweight LLMs.

LFM2.5The next generation of on-device AI
shemith mohananleft a comment
Love the idea of an ultra-light, distraction-free eReader. The magnetic back + pocketable form factor is a smart touch—feels designed for actual daily carry, not just desk use. Curious how the paper-like experience holds up for long reading sessions. Congrats on the launch 🚀

Xteink X4Ultra light, magnetic E-reader that attaches to your phone
shemith mohananleft a comment
Data quality really is the hidden bottleneck for AI. Interesting focus on stability and compliance — reliable long-term pipelines matter much more than just raw proxy access. Curious how teams are using this in real-time AI workflows.

ThordataFuel AI training with high-quality, scaled data via proxies


