I'm considering hiring a coder and I was wondering what peoples' experiences were with this and what site and/or specific people they could suggest. Specifically, I'm looking for someone to code a platformer backend/library for me including slopes and character collision. The most comparable game would be something like Guacamelee [video]. (I've tried physics libraries like Box2D and while I love them, they're not a good fit for this type of gameplay.)

Any suggestions on where to go for this? Thanks!

Craigslist?

My only recommendation is to realize this is going to cost a lot of money and take a lot of time.

You are turning down engines that have many thousand development hours behind them, and asking for something custom.

Even with a bare minimal set of functionality you are probably looking at hundreds of hours. You're probably looking for months of work at least, which means paying someone for months of work. Think about a reasonable pay for a programmer capable of what you want, multiplied by the hundreds of hours it takes.

Thanks. Unfortunately, as far as I know, no engine does this in the way I showed above. The closest I've seen is the Corgi Engine, written exclusively for Unity. This is the most likely possible alternative short of me trying to code this myself.

And yeah, this wouldn't be cheap though I estimated dozens of hours just based on how long it took me to make a tile-based platformer.

Unfortunately, as far as I know, no engine does this in the way I showed above.

What way? All you mentioned in your original post was a platformer with "slopes and collision," which are absolutely, 100% in the scope and capabilities of Unreal, Unity, and nearly every other game engine I can think of.

What do you propose to be doing differently?

Specifically the Guacamelee link I included. Back in 2013 I contacted Drinkbox, the makers of Guacamelee, about if they use Box2D. This was their response:

The answer to your question is actually pretty muddy. In a nutshell, we use Box2D within engine, and not have Box2D as our engine. Our first titles (main character using treads) used Box2D more in the actual platforming, but for Guac, too much control was demanded to let Box2D actually dictate all the movement. I think this is why you might not find a lot of resources that have Box2D as the driving force behind a platformer. For Guac we mostly used Box2D to drive the effects - for example - the way body parts of an enemy flew through the air. Box2D is great for this, but, not so hot to drive the character controller.

As a small example, I use Leadwerks. It has a great character controller but it's designed for 3D. To make it use 2D you need to start jumping through hoops. Even then, if you run down a slope fast enough, you start bouncing. Which makes sense as far as real-world physics are concerned but not so much for platformer games. This is the type of stuff I want to avoid by messing with existing engines/libraries.

This is not a problem you solve by reinventing a new engine. This is a problem you can solve within any existing engine by doing essentially the same bits of logic and character controller work you'd do writing your own engine from scratch.

Any programmer you find who is skilled enough to implement what you want should be skilled enough to do it in an existing engine. If they can't, they probably aren't actually skilled enough.

Further, by doing it that way you'll save yourself probably tens of thousands of dollars.

To be quite frank, it might not be wise to making these decisions on what can/can't be done with certain tools. I don't know what level of programming experience you have, but if your looking to outsource this, i can tell you it wouldn't be overtly difficult to override the way box2D handle's things(in the video it looks like the biggest thing they did was not allow the character box's to actually rotate along the slope). It might be better to ask how the person your outsourcing would accomplish the goal rather than trying to make that decision yourself.

I echo everyone's sentiments here in regards to using an engine. I don't think we need to beat that dead horse anymore.

I was thinking about how I would interview a coder while I was taking a shower. I think the very first thing I would do is introduce myself, talk about the company, explain the position, and then ask the candidate to solve a few programming problems. The first would probably be the "fizz buzz" test (and maybe changed a bit to prevent study by memorization), where you weed out the people who can't code whatsoever. The second test would be to create a function which can find any point on a circle, given an angle. This should be absolutely trivial, but is designed to make sure they know some basic trig. then, we code another function which can find any point on a spiral. This should be moderately challenging, but not impossible. The last test is to give them a specification for an RPG game and ask them to design / architect a class hierarchy. How do they plan out object inheritance and relationships?

If they get through these gates, then they've shown that they have some technical chops and its then worth looking further into their past experience and personal values to see if they'd be a fit for the culture you're trying to create.

I would also add that the scale of such work is probably more suited for partnership rather than a for-hire programmer, just simply due to the length of time and amount of work need to be done.

If I were you, I'd pick an engine that you can be comfortable of using (e.g Unity), then hire a programmer to build a template platformer game for you in that engine. Single-level, with all the slopes and collisions you want. This is more reasonable to pay a programmer for. I wouldn't be surprised if there is a programmer out there who has already had such a template that he/she can just hand it off to you with small brush ups and adjustments.

This does not mean it's the final version. There is still a lot of customization and work needs to be done to make it an actual and complete game. This is the work that you should partake yourself. So, you'd still end up doing the coding yourself, but at least much of the core work are already done by someone else.