High surface temperature would be a problem, but far from the only problem:
- Atmospheric pressure: the atmospheric pressure on the surface Venus is 92 times greater than on earth. That has several implications:
- Rocket nozzle expansion: Rocket engine bells work best at the pressure they're designed for, with bell expansion being much larger at vacuum than at 1 atm. 92 atm to 0 atm is such a wide range you're probably going to be forced into using an aerospike, which is perfect for this application but a large aerospike has never been flown so you'll need to develop this.
- Tank pressure: Launching from 1 atm fuel tanks have to be pressurized to maintain tank integrity as the fuel is pumped out. On Venus you'll have to maintain the pressure high enough that the 92 atm of ambient pressure doesn't crush your tank as you lift off. This adds extra weight and complexity to your rocket.
- Drag: The atmosphere is thick enough that the "soft landing" of Venera 13 involved releasing the parachute and just falling at terminal velocity for the last 50 km (with an airbrake, but still). Your rocket needs enough thrust and delta-v to push its way back through that.
- Maximum dynamic pressure: 92 atm pressure at the surface doesn't tell the full story. During ascent the rocket engine is pushing up and drag is pushing down so your rocket has to be structurally strong enough to not be crushed.
- Gravity: Acceleration due to gravity on Venus is 0.904 g (90.4% of Earth's gravity). That means that if the atmosphere were similar you'd need to take a fueled rocket capable of escaping Earth with your desired sample return mass and land it safely on Venus. As noted above, the atmosphere on Venus makes for a much more challenging environment than Earth so you need a bigger rocket on Venus.
- Temperature: Electronics don't like heat. The Venera missions found ways to overcome that for a short mission with a heavily insulated chamber, or maybe you can build electronics with high heat tolerance. Any gasses in your rocket are going to want to expand as they heat up, and you should make sure your rocket fuel doesn't boil below 500°C. Having hot fuel is going to impact the regenerative cooling of your rocket engine too, so design extra temperature tolerance or another solution into it.
- Atmospheric composition: There are sulfuric acid clouds but the atmosphere in general is 96.5% carbon dioxide. Fortunately, dry CO2 isn't corrosive so this might be the least of your worries.
- Wind: You're starting off with winds of 0.3 to 1.0 m/s at the surface ("but due to the high density of the atmosphere at the surface, this is still enough to transport dust and small stones across the surface, much like a slow-moving current of water") and getting to winds of 100 ± 10 m/s in the upper troposphere. So long as the winds pick up gradually without any sheer breaks this is probably manageable.
Most or all of the above factors will add complexity to your mission, and any additional insulation, structural supports, shielding, etc will add mass. Added mass means you need a bigger rocket to leave Venus. A bigger rocket to leave Venus means you need to get that rocket fueled and on a transfer orbit from Earth to Venus, meaning you need something capable of launching your Venus rocket from Earth.
Something else to keep in mind: As far as I've been able to find the only things we've launched separately and assembled in orbit have been related to manned flight: space stations, Apollo-Soyuz, some flights of Project Gemini, Hubble repair missions, etc, never multiple unmanned pieces of a probe to be assembled in orbit. That's not to say it can't be done, just the added complexity is significant enough nobody has tried it for a probe so far.
Everything in the challenges section assumes dropping a lander all the way to the surface and using rockets to get back into orbit. Building off the responses to comments:
- Use an aerostat to keep from dropping all the way to the surface and lower a cable. This probably has a lot of promise for something like a colony that has more infrastructure in place (the context of the paper that suggested it), but I feel like a 50km cable (which keeps you at 1 atm pressure) is a bit much for a sample return mission. Also, the winds "have a strong vertical gradient" wikipedia so without a heavy weight on the end you'll end up trailing the cable more horizontally than vertically due to drag on the cable.
- Use a rockoon type design: drop to the surface, collect your sample, then use a balloon to lift you out of the more dense atmosphere before firing a rocket when the balloon stops rising. You should be able to get a large amount of buoyancy from a balloon considering the density of the atmosphere but I don't know what challenges you'd face trying to fill the balloon and keep it from popping on ascent. For reference, a weather balloon (from 1 atm to >0 atm) expands by 83 times by volume.
If you're familiar with Kerbal Space Program look up some of the monstrosities that people have built to return from the surface of Eve. That has some similarities with Venus, but doesn't have to deal with temperature or wind and uses a simplified model in many areas.
I know I missed stuff. I know there could be more numbers included to give a better sense of the scale of the challenge. Any edit for the sake of adding data I didn't include is welcome, or add a comment and I'll try to incorporate the data for completeness.