Please do! I'm here for the expertise of those already well versed in the game.
Yes, well, I could talk about Inquisitor's game design until the grox came home. I've quite possibly spent more time thinking about Inquisitor's rules than anyone else, including the original designers.
One of the aspects I'm curious about is how to appropriately scale threat to the number of characters. I'd like each encounter to be challenging, but not quite push the players to despair.
Well, at its core, Inquisitor was designed as a PvP game. While it is GMed, the GM is often not the antagonist; while he sets the scene and adjudicates things, it's entirely possible for games to have no NPCs at all.
More often, the primary antagonists for the players will be each other, different Inquisitors (or other infamous personalities) taking their own sides in the Battle for the Emperor's Soul.
This makes GMing be a little unusual compared to more conventional RPGs, but it is actually a very refreshing experience.
My games of Dark Heresy are often fairly predictable to gamesmaster for - I know the general course of the plot from the start and while the players will certainly deviate from that somewhere*, it's all going to end up broadly where I expect unless I inflict a TPK**. * The previous plot I ran involved an unexpected shopping trip for dresses, but still, the players did eventually work out who the bad guy was.
** I never have so far - my players are in it for the story, and "everybody dies" isn't usually a good story. While characters do die occasionally, if it looks like a fight is going TPK badly, I'll usually let the players try something creative or have the bad guys take them prisoner instead.
With Inquisitor though, any of the players could succeed, meaning even as a GM I am discovering the game as it plays out. Even really simple game concepts can result in outcomes I never imagined.
One of the best games I've ever run just started with the instruction that the players had to work out which of the three of them had sabotaged the plasma reactor they were running away from. (I'd secretly told one of the players he was responsible; he was supposed to escape without getting found out, the other two were supposed to catch them). All three of the players got really into it, coming up with creative ways to extract confessions, spot clues or misdirect blame on to each other; in the end, the culprits were discovered, but chose to be taken for trial rather than try to shoot their way out.
Of course, that's not to say that the GM never plays NPCs or that the players are always enemies. The Inquisitor rules are versatile enough that they can be made to work as a PvE system, but their real unique selling point is for this PvP play.
As I've already sort of mentioned, this shows up most in Inquisitor's damage system. To newcomers, it can seem quite harsh, as it's built on the expectation that players will have more than one character (allowing them to persist even if one of their characters is incapacitated), and also quite detailed, so that players can hinder their opponents without just putting them out of the game. The need for this detail is aptly demonstrated by Knight House armies in WH40k, which are often criticised for being frustrating to play against - you either feel like you're doing nothing, or you take a couple out and it's suddenly really easy.
Having that detail in Inquisitor's damage system makes a player's hits means something, but without finishing fights boringly and unheroically fast. It's not fun for any player if a character goes down too easily.
Naturally, there's still a question of balancing players' characters to each other. However, with some of us having been playing this for 15 years now, there's quite a lot of guidelines and advice on that.
The original random stat generators have rather fallen out of favour - we tend to prefer hand written profiles. Players are much more intelligent than dice (normally...), so they can usually do a better job of writing fair profiles that accurately represent the characters.
We've built up the "Conclave Standard" over time from general experience of what works best. (A link to one of my write-ups on that
But generally, it's just about having the right attitude. There's no point in winning by bringing the hardest characters (that'd be like bringing 3000 pts to a 2000 pt 40k game), or really in "winning" at all. It's all about playing fantastical characters and having cool stories - it's not about the bragging rights.
And there's nothing to be ashamed of if you do get the balance a bit wrong. If a character doesn't work quite how you intended, go away and change them. Even if you get it catastrophically wrong, the GM is omnipotent - things can even be fixed mid-game (it's a lot better than persevering with something that isn't working).