the hardest thing about living in SF during the 2010s was that the wealthier it got, the worse it got. the opposite should have been true.
it could've become this
Today @getspara launched Scout, our AI assistant for sellers!
I was the tech lead and it’s been a WILD ride.
I went in with a pretty strong idea of what tech leading looks like. I had to rethink most of it on the fly.
Scope
I used to spend a lot of time negotiating what we could fit into a project. On Scout, we could take on a lot more at once. Tech leading in the AI era means managing an onslaught of moving parts. Psychologically tempering this, staying organized and addressing problems on the spot become paramount.
Complexity
Scout ended up both wider and deeper than what we could’ve built pre-AI. It also evolved faster than I could keep the whole thing in my head.
The tech lead should know as much as possible about the thing they’re shipping. I leaned on AI really heavily here. I could go from “how does this work?” to “is it healthy?” to “are sellers actually using it?” in one sitting (thanks Claude Tag!).
Role
Tech leads were responsible for derisking the technical component of a project pre-AI but now as well. What derisking involves is way different now. For us, this often meant ensuring what we built was correct from a product standpoint, as opposed to a technical one.
Project management
I used to throw the book at this. Firm scope, careful schedule, constantly reprojecting dates. On Scout, our roadmap went a week out at best. Sometimes it went a day out. The team gobbled through it.
The team’s been laying the groundwork for Scout all year. I’m really proud we got to put it all together and ship it today!
Link below in comments👇🏼
Today we’re launching Scout, the AI assistant built for sales.
It lives in your Slack and text. It works every account in your book and tells you who to talk to, why now, and what to say.
Then learns how you sell and drafts the next best action to take in your voice: an email, a LinkedIn message ready to send.
We ran it on our own outbound first. Reply rates went up 10X.
The future of sales is here, and it’s about to close you a lot of deals.
Request early access:
https://t.co/Koi7kngIJo
I've watched this team generate an incredible amount of momentum to bring this product into existence. Witnessing this has been the most fun I've had at @getspara, hands down.
If you’re in sales, this is for you! Check out Scout and get on the waitlist ASAP. Link below 👇🏼
Today we’re launching Scout, the AI assistant built for sales.
It lives in your Slack and text. It works every account in your book and tells you who to talk to, why now, and what to say.
Then learns how you sell and drafts the next best action to take in your voice: an email, a LinkedIn message ready to send.
We ran it on our own outbound first. Reply rates went up 10X.
The future of sales is here, and it’s about to close you a lot of deals.
Request early access:
https://t.co/Koi7kngIJo
@kunchenguid To me, this means that the edge of creativity has been pushed further out. That's a good thing!
There are an unlimited number of things you can type into an LLM, and I can guarantee you that humans will keep seeking out and finding creativity here.
Code review has really become, "did my colleague type the right thing into Claude?"
With software development happening as fast as it is now, there's less time to define product requirements up front and more winging it, and review is the logical checkpoint to ensure that at least you're aligned on what was supposed to get built
maybe a hot take but i now consider adversarial reviews mostly a waste of tokens. far better to use those tokens on empirical combined with formal verification. i have updated pstack to use the interrogate skill less, and eventually i think i may remove it altogether
i have found very little value with code review in the abstract, agents tend to just get lost in the nitpicks and yagnis