Head of Product @ TinyFish. Former Product Lead of Tableau AI & Agentforce @ Salesforce.
product person. coffee lover. a builder.
new on X. follow the ride~
As much as I appreciate your effort here to create noise, seems like you missed the most obvious signal. Give your agent a spin with @Tiny_Fish and tell me if we’re wrong (hint: you can actually use it inside Monid too). Developers like yourself care about cost performance above all, and there’s no one else who does it better than we do. Keep us honest here and let me know if you need any help:)
The TinyMerch Challenge is live!
It's only open to members of the TinyFish Student and Ambassador Programs.
Design TinyFish merch and watch the votes roll in live on https://t.co/4hculKn9VA. The winning design becomes a limited TinyFish drop.
Go design now. Go design now. Claim the TinyMerch Challenge on TinyBounties and submit your design.
Entries close Sunday, 15 November.
What a fun time at @hyperspell’s office! They’re building the company brain that actually understands and does the work. Thanks @contextconor for making me feel at home with such a warm and special welcome!
Context matters more than ever now. Winners will make the collection and consumption of context truly frictionless and deeply embedded into every flow of work. Excited to see where this goes🔥
Make sure to check them out: https://t.co/mHYkGxuIKD.
@CoreSpeedHQ made this using their video studio with Opus 5.5 after integrating with @Tiny_Fish. Thought I had to share for the 💯 creativity 😂
Follow for more
A customer-facing agent can work perfectly in a demo and still be nowhere near ready for production. We had a good conversation about this at our SF Tech Week session on building and deploying customer-facing agents, with @simonrohrbach, @hey_amandam, and @Homer_Wang.
A few things stuck with me.
1. Calling your agent once and seeing it work isn't testing: Amanda made this point well. You need evals around specific decisions, simulated conversations with different customer personas, and monitoring once agents are live. You can't manually call every agent every time something changes.
2. Your agent is only as good as the context it has: Simon Plain shared something interesting about knowledge freshness. Human support teams can give inconsistent answers. Documentation can go stale. The codebase often gets you closer to how the product actually behaves. And even if your knowledge is accurate today, your agent can drift as the product, models and human behaviour change. Getting the right answer is more than just retrieval. It's about knowing when your source of truth has changed.
3. Making the web accessible to agents shouldn’t come at the expense of the web itself: The web needs to be accessible to agents, but access can't come at the cost of overwhelming websites or ignoring their boundaries. There's a balance to get right.
A few good conversations with people building this stuff for real. Thanks to everyone who joined us!
ICYMI: Hermes can search and read the web for $0.
https://t.co/Z0B83N8JG7
TinyFish is now a plugin in the @NousResearch Hermes Agent.
We pointed ours at the most boring job in the building: an Excel sheet full of empty cells.
TinyFish searched and fetched information for all the rows and handed back clean results to Hermes Agent to fill in the spreadsheet.
Muse gave your agent a computer.
Instinct gave it your phone number.
We gave it eyes.
Introducing NewEyes AI — your personal agent with eyes.
> Glasses see it.
> Earphones say it.
> It remembers where you left off.
> "Good morning. Continue on page 212."
Your world is the context.
Had such a great time with @simonrohrbach , Homer Wang, and @ravibits talking about all the ways we approach getting agents to production.
Thanks to @scalekitinc , @plainsupport , and @Tiny_Fish for having me and to a truly engaged audience with challenging and interesting questions
@latentspacepod has such an incredible studio down in South Beach! @swyx - @realbasilchatha is the real one to get it done. Basil and I had a ton of fun exchanging on what we’re working on and where web+AI is heading. Can’t wait for the production to be out soon! 👏
What a cool event from @WorkOS! Thanks for the invite team! And shout out to the amazing @grinich!! Simple, to the point, great agenda. Best time to build agents that can navigate the entire world, with trust and security in mind.
I also loved the popcorns and lemon-aid! ☝️
#init
We set a goal of 100,000 users by October. Today we crossed 300,000.
The first 100K users took four months.
18 days to 200K.
10 days to 300K.
Every milestone came faster than the last, and it held up because of months of work our team put in before anyone was watching:
https://t.co/mesJNIzkSm