It's fairly simple. You write values using keys and expiry times. You get values using keys. You can expire keys from the system.
Most clients follow the same rules. You can read the generic instructions and best practices on the memcached homepage.
If you really want to dig into it, I'd look at the source. Here's the header comment:
"""
client module for memcached (memory cache daemon)
Overview
========
See U{the MemCached homepage<http://www.danga.com/memcached>} for more about memcached.
Usage summary
=============
This should give you a feel for how this module operates::
import memcache
mc = memcache.Client(['127.0.0.1:11211'], debug=0)
mc.set("some_key", "Some value")
value = mc.get("some_key")
mc.set("another_key", 3)
mc.delete("another_key")
mc.set("key", "1") # note that the key used for incr/decr must be a string.
mc.incr("key")
mc.decr("key")
The standard way to use memcache with a database is like this::
key = derive_key(obj)
obj = mc.get(key)
if not obj:
obj = backend_api.get(...)
mc.set(key, obj)
# we now have obj, and future passes through this code
# will use the object from the cache.
Detailed Documentation
======================
More detailed documentation is available in the L{Client} class.
"""