
AI Engineer @howie_ai. Previously: Founder of Elephant AI, SE @GetYourStake, SE Zuma AI, SE @Workday, MD @USCLavaLab. CSBA @USC.
Seattle, WA
Joined March 2011
- Tweets669
- Following1.6K
- Followers484
- Likes4.7K
What I’ll remember most about the last 2+ years is how many times we rebuilt Howie. We never treated what we had as good enough, so whenever it hit a ceiling, we went back to the drawing board. Relentless focus on the same problem.
We decided early that code is worthless. What matters is what you learned writing it. That only works with leaders who do the work themselves. Thank you @whatupdave and @awwstn.
To the rest of my colleagues at Howie, your dedication made this possible. I'm grateful for the opportunity to continue working alongside you.
We’re excited to welcome Howie to Google Labs!
The @Howie_ai team has spent years figuring out scheduling and calendar management. We can’t wait to learn from them as we build the next generation of intuitive AI tools together for small businesses 🤝
I built the agent that decides when NOT to respond i.e. the part that does nothing
The Certifiably Insane Way to Build an AI Agent:
1. choose a category where mistake tolerance is roughly the same as it is in self-driving cars. we chose "email-based scheduling assistant." many people want this product, but they immediately fire him if he screws up an interaction with a prospect, a candidate, or a potential investor
2. you learn that the edge cases are too complex and too frequent to be solvable. ours: managing timezones for people who travel (and change travel plans) constantly. knowing when NOT to respond, when to text the customer on the side to verify something, when to follow up, which sub-calendar to use, when to bend the rules on availability, when we can schedule that one type of call during your commute but not the other type of call. sharing your availabilities without compromising your privacy. and on and on.
3. the product doesn't feel viable, but you don't want to give up. you spend hours in a hot tub in Marin with a friend who makes self-driving cars. you make a plan to do it the way they did: hold the steering wheel. you go home and build a human-in-the-loop platform and hire contractors to serve as a backstop and catch mistakes before they happen (and to help design a map of what a world-class EA would do in every weird scenario). you decide trust is the currency in your category, so it must be the thing you won't compromise on. the product must succeed at any scheduling request, no matter how complicated.
4. you instantly feel an overwhelming market pull. so you keep going, growing that team to 75 people working 24/7 to support the nonstop scheduling needs of your customers. tons of engineering time goes to scaling the human platform instead of building the product.
5. you try to raise a Series A and investors say you are insane. your gross margins are extremely negative. they believe this is a problem worth solving, but they don't believe it is as hard to solve as you say. they want AI, not humans. your competitors put "NO HUMANS IN THE LOOP" on their landing pages to call you out. you keep going.
6. you work day and night building the harness that can meet the quality standard your customers have come to expect. you create a massive synthetic gold dataset. audit it, and clean it, label it. repeat. then, experiments. fine-tuning. RL. ACE. DSPy. sub-agents. sub agents for your sub-agents. rebuild the harness. throw more tokens at the problem.
7. some weeks you make big progress. some weeks your evals climb a single basis point, but that's better than nothing. more experiments. more tokens.
john coogan said the hot trend in 2026 will be dogged pursuits. that pushes you to continue the pursuit, doggedly.
8. then, one day, you realize you are scheduling thousands of meetings a day and approaching 50% autopilot with no increase in churn or complaints. you put 150 customers in a full self-driving experiment, and they use the product MORE than they were using it when they had the human backstop. you can really start to let go of the steering wheel.
9. you don't know yet if this was a hill worth climbing, but you are nonetheless stoked that you can see the top. you have created a proprietary map of what to do in a million different situations. nobody else has that map, and the models keep getting better at following maps. your plan was to bet on trust, and your product can be trusted.
today was the first day Howie crossed 50% autopilot:
thanks austin! this is just one of the many things you can put on autopilot when you hire @howie_ai as your EA
it's actually pretty hard to get a prompt to write a useful tweet-length summary of coming days of your calendar, but @_connorduffy has nailed this
1
243
My 2024 Wrapped 🔥
- rejected by YC
- launched my project management agent as an iOS app anyway
- sold my IP to @howie_ai
- started training an agent for automating scheduling with a human touch
6
400
Connor retweeted
introducing: The Internet History Calendar
for each day in 2025, you can relive a fun and interesting moment that occurred on that day in internet history
links to the free Chrome extension and the $20 physical calendar in the next post.
Over the next 5 weeks, I'm going to ship Elephant AI to the App Store. You can join the beta at the link in my bio.
dm me if you have any feedback.
cc @_buildspace
1
2
284
I was just accepted to s5 and now I've got a buddy pass to give away for @_nightsweekends dm me if interested!
2
196
When I first started hacking LLM apps, I bootstrapped together a voice assistant like we just saw in the OpenAI spring update. I realized that whoever solved latency would win, and I'm thrilled to see they have accomplished that with GPT-4o.
2
282
gpt-4-turbo-preview likes to annotate it's response as markdown if you ask it to use markdown formatting, so I just built this little util function to manually remove the annotation #buildinpublic
1
247
Connor retweeted
startups try too hard to fire unimpactful people and they don't notice all the negatively impactful people
I'm seeing great results programming GPT-4 to have structured access to highly technical things (like code documentation), and asking it product questions.
3
319
Switching to Safari when I don't need extensions has been life changing. I'm just air dropping links from iOS Safari to my Mac. Significantly faster for the most important action on mobile.
Bonus points to @Apple for taking up way less RAM than Chrome and making the "Empty Page" on new tab easily configurable. Looking at blank html helps me think more clearly.
1
373















