Topic title speaks for itself… I hope… Personally I had a flurry of expanding my setup with lights, sensors, various hubs, (for me) writing code… But after all that…. Sigh… Another exhale…
How have other people found a second, third (or other) wind…?
To me it feels like once HA projects dwindle and life gets in the way it is hard to find a way to get HA back up the list…
For me, it was AI.
I'm not a developer—I'm a retired Project Manager who had the privilege of working with some incredibly talented developers over the years. I picked up enough to be dangerous... emphasis on dangerous. 
Once my automations reached the "don't touch it, it's finally working" stage, I hit the same plateau. Then AI came along.
I've been using it to resurrect abandoned drivers, understand code that once looked like ancient hieroglyphics, track down bugs, modernize old projects, and even add a feature or two. Some of my ideas have worked surprisingly well. Others have resulted in me asking, "Well... that seemed like a good idea at the time."
I've also learned that AI is very confident—even when it's completely wrong. It'll happily suggest a function that doesn't exist or invent an API call out of thin air. So there's definitely a healthy amount of trial and error (and occasionally a few words not suitable for a family forum). 
The nice part is that it has made the hobby fun again. Instead of just maintaining what I built years ago, I'm learning new things, revisiting old projects, and breathing new life into drivers that might otherwise have been forgotten.
So while I still wouldn't call myself a developer, AI has become the incredibly patient coworker who never gets tired of my "one more question..." moments. It almost feels like having one of those great developers I used to work with looking over my shoulder again—except this one doesn't mind explaining the same thing five different ways until it finally clicks.
Note to self…. Include @DGBQ in “the group”…. 
Thanks David, while I’m one of the group who taught you the dangerous arts of a dev, plus the ills of AI.. Not sure I can taint you all with too harsh a brush….
But seriously, I can understand your POV… But am too tired to type out the remainder of my reply… Needless to say that you are likely on the right track…
I hit the plateau a while ago. Getting Emulated Hue working in Home Assistant, with Hubitat devices passed to HA via Maker API, so I could get some of my lights controlled by my Harmony remotes was a major highlight in home automation for me. I pretty much have everything I need and it's all working as I need it to. I've since integrated Apple Home and have added/removed/upgraded devices such as locks and thermostats, but that's been the extent of it. That is until I received my SofaBaton X2 remote yesterday.
This remote is fantastic, and as much as I still love my Harmony Elite, the Home Assistant integration is a game changer for me.
I've added the MQTT Broker to Home Assistant, along with the SofaBaton integration, and have been adding automations and programming buttons and macros all day.
I have reached peak WAF home automation until MMWave occupancy sensors became a thing and got the Inovelli wall switch version. It has made room resets work so much better when the room is unoccupied. All other experiments with lux and other sensors has been one off’s and don’t directly effect the occupants of my home just do maintenance items.
With AI and unpredictability I have no intentions of having it in the loop for home operation. I have used it for cleanup of some custom code I wrote for my weather station integration and voice announcements using the now dead lannouncer android apk. And testing some off grid control with LoRA Meshtastic/Meshcore functionality.
I have not plateaued. My recent incorporation of HA, imported via HADB, has enabled a high temperature wood stove sensor that I had searched in vain for years for HE. Integrating AirThings sensors and a couple Bluetti 180's, also via HA and HADB has been nice as well.
All the new features coming out on HE don't apply to me, but the (fairly) recent HomeKit integration and my (fairly) recent adoption of Apple world has allowed the HomePod minis to be a very nice voice announcer. Although, I might change the default guy to a more agreeable lady. 
After spending hundreds of dollars on a failed Sinope oil tank level setup, and failure of HE staff to get the sensor working to any degree of accuracy, despite sending in the level sensor, I'm trying to get HA to work with the 433mHz radio on my existing Beckett Rocket wireless gauge. Muddling along though-maybe not possible...for me. Bought the DVB dongle and reading up on RTL_433.
I don't know if I will ever delve into AI coding. But never say never.
I plateaued with Hubitat about two years ago, when I discovered Home Assistant. I’ve largely plateaued with it as well, but taking up gardening in a big way has caused me to fiddle with ESP32 devices and soil moisture sensors (including the Ecowitt WH51). Installing an air-water heat pump has helped renew interest in tracking and controlling things as well. I find I need the tech to DO something to keep me interested.
I’ve more or less plateaued with HE and recently stuck my toes in the water with Home Assistant. I’ve got a few things running on it that were unreliable (or using unsupported, abandoned or buggy implementations) on HE, that are working nicely on HA. Best of both worlds.
Setting up HA wasn’t nearly as scary or difficult as I thought it might have been and the HADB bridge works great.
This is the same for me. Hubitat is great but HA lets me sort some rough edges out. I have had HA up and running for 3 months and am very pleased how it compliments my Hubitat setup. HA allows me to bring in my Eufy camera alerts and arming and also automations using my BYD as a trigger.
We just moved for the 3rd time in 7 years. It is not the ideal way to keep things fresh but it makes the cost of tech look small by comparison.
The new place adds some new opportunities. HVAC is 2 zones, which is new to me. The pool is ripe for automation with a solar heater and electric spa heater. A water fountain pump to schedule and maybe add an auto-fill feature. Lighting and fan upgrades are standard but a new configuration and flow. Plantation shutters are begging to be replaced by motorized blinds. Irrigation system to upgrade. Two low voltage lighting panels that have as-built labels that probably haven't been accurate in over 20 years.
I can take a beat and consider how to harmonize choices with bigger consequences - like adopting Lutron Caseta and sacrificing some powered repeaters for Zigbee and Zwave.
There's always something new just when you think you've done everything. I tend to have 'I wonder if I can do that' moments; then I'm pursuing an idea for something that quite frankly isn't necessary. An example of one of my recent brain farts and how it started:
1 - Watching Clarkson's Farm on Amazon Prime and noticed the bird song identifier app he was using (Merlin). Thought 'That's clever' and downloaded said app.
2 - 'I wonder if I could do that in the background with some app for Pi' - Found BirdNet Go and installed. Worked out I could use RTSP audio from my existing Hikvision cameras so configured that
3 - It has MQTT so I configured to publish the events to Mosquitto and then node-red to capture the JSON.
Next possible steps - all unnecessary and I'm only doing it to see if I can:
1 - Alerts to dashboards of what is chirping away outside. Need to get node-red to assemble a string with time, species, and confidence percentage
2 - Some sort of page for dashboard showing the days birds identified
3 - Have the currently singing bird identified on one of the configurable text overlays on the cameras that are capturing the audio
Cool if you're somewhere with a lot of bird activity but really unnecessary!
Somehow I seen to have produced a group of “HA advocate” topics… hopefully I won’t get banned… or those who have contributed…
I certainly wasn’t setting out to make this (or other topics) focused on alternative automation platforms, but I am glad people have found some benefit from broadening their systems to encompass more than just one platform. I do still feel that options are an important element in any smart home.
It feels like I might be regaining my motivation, for multiple reasons, but including peoples responses here and on other topics….. Always great to hear other people’s stories and take inspiration from them.
Based on my last comment, plus @Timbb‘s “rough edges” comment, it makes me think that external platforms are servicing that purpose at the moment, which in my mind is fine… for now… but can also appreciate that Hubitat are making an effort to bridge the gap.
And who knows… I may exit my plateau and restart my Tasker, CSS and other projects…
I feel you. But at the same time I feel great about it.
My WAF is super high, I’ve used AI to fix, add and migrate off of some things i was too lazy to do before, and I couldn’t be happier. My wallet is happy I have slowed down on buying new devices, and I figure I’ll jump back in hard when either something breaks or there is a new thing to integrate or automate!
I’ve peaked.
Just use AI to vibe code and look at updates i should make to my own rules/code.
While i do love Inoveli, the mmWave are a bit outside of my price range to start wanting to try them out. With switches, you like one you then need to change them all.
I always like the ‘ideas’ forum to get new ideas, since i’m not as original (or just lazy) to think of my own, some here get pretty techical.
I don’t think you have anything to worry about. The HE folks have, in my opinion, always been pretty open-minded and tolerant when it comes to discussions that involve other automation systems. No one system can be all things to all people.
I’m a novice HA user, but it seems to me that I could ditch HE and move everything to HA if I was so inclined, but I’m not. I have four HE hubs and no intention of getting rid of any of them in the foreseeable future. HE does many things very well, and for the few things it doesn’t HA fills the gap. If HE ever became unviable for some reason (hopefully never), then there’s always a plan B.