How do you solve a problem like AstroScript?

Very Similar to my build but I think your's is much more straightforward :) I think I'd like Medium more than Keyhole here but other than that it looks insane.

I was trying a very similar build with Scavenge & Imp recursion instead of Keyhole, and Emergency Shutdown instead of Account Siphon. However, I like your build a lot more since Vamp alone was definitely not enough for pressuring the corp in mine.

Did you experience corps icing HQ heavily after the first siphon/vamp, and if so, was Parasite enough to open the way again for subsequent siphon/vamp/nerve agents? I ask because that's why I ended up going the Emergency Shutdown route in the first place, but I was never happy with leaving out the Siphons to free influence.

Also, I think I'd run Deja Vu over Same Old Thing every time. It's more versatile and plays using less clicks. I'd rather have an extra click than 2 credits most of the time. Recurring Parasites is a thing in many matchups and it can still recur Account Siphon in a pinch.

I think I agree with swapping Deja Vu for Same Old Thing. I've played two games with this version now, and although that is by no means sufficient testing, I've ended up discarding Same Old Thing every time because it was such a large tempo hit, and drawing more cards was always a better bet because 2/3rds of the game was spent searching for the appropriate breaker anyhow. Here are the changes I will test next:

-3 Same Old Thing +2 Deja Vu +1 Datasucker (helps getting one earlier, duplicates do not hurt, good target for Deja Vu if you lose one to net/brain damage early on while facechecking).

Since there are no tutors, having 3x of more cards also makes the deck more reliable (and also less flexible, but we'll gloss over that detail for now). What do you think?

All good thoughts :) I should probably point out that since the tournament I've changed to -3 Cyberfeeder, +1 Deja Vu, +1 Keyhole, +1 Grimoire (Feeder wasn't giving me credits soon enough in most matchups). I haven't found I've needed a 3rd Datasucker since I've mainly been facing small ICE decks like NBN, NEXT and Jinteki, and Knight tends to handle big ICE well. I may include a third Deja Vu, but I would want to keep 3 SoT (it's a little cheaper, which can matter a lot), so I'm not sure what I'd cut. While it can be a tempo hit, it's not as much a tempo hit when it's recurring QT or Siphon, since they give you the tempo back.

In terms of heavily icing HQ, the three Anarch Core Set breakers will get you almost anywhere, with Knight picking up the slack. I was lucky to not see a single Tollbooth all day, but Parasite will just be killing most big ICE if the rest of it is small, because your Datasuckers will just pile up. I was more common that they would heavily ice R&D after Keyhole, then I would Siphon.

Since my last comment I've gotten three more games with my version (the current list shows up under "Inspiration for"). I agree that Cyberfeeder rarely gives the credits in time, but in my experience it was a quite convenient way to protect the breakers from Power Shutdown. Granted, maybe it is not worth playing just to get around Power Shutdown, and it doesn't protect Datasucker at all, but I had come to like the feeling of (false?) security. Barring the Deja Vu/SOT choice, I do agree that Cyberfeeder would be the first card to cut if I wanted to add something new (say a new card from Honor & Profit that we don't yet know about).

As for the SOT vs Deja Vu, I agree that SOT is cheaper, but since it's only cheaper by 1 credit in comparison to playing Deja Vu+clicking for credit, and they are equal in cost if Armitage is installed (of course Armitage is not endless, there is that), I like the additional flexibility that Deja Vu provides for obtaining breakers (mostly Knight), or taking two viruses (mostly Parasite).

Adding another Keyhole & Grimoire is very interesting too, as I'm always happy to see both early, and like I said I like having 3x cards in the deck for consistency, but at the same time the second and third copies of either are rarely useful, and they are not as critical to the game plan as the breakers. I am divided on this one, but I would very much like to hear about how it works out once you've had a chance to test the changes.