Generate N random numbers fulfilling some criteria; see https://stackoverflow.com/q/61352556/1968

simple.r

alpha = numeric(1e4L)
beta = numeric(1e4L)
i = 0L
while (i < 1e4L) {
  a = rnorm(1L, 10, 2)
  b = rgamma(1L, 8, 1)
  d = a - b
  if (d < 1) {
    i = i + 1L
    alpha[i] = a
    beta[i] = b
  }
}

r = data.frame(alpha, beta)

vectorized.r

alpha = numeric()
beta = numeric()
i = 0L
chunk = 20000L
while (i < 10000L) {
  a = rnorm(chunk, 10, 2)
  b = rgamma(chunk, 8, 1)
  d = a - b
  valid = d < 1
  alpha = c(alpha, a[valid])
  beta = c(beta, a[valid])
  i = i + sum(valid)
  chunk = max(chunk %/% 2L, 10L)
}

r = data.frame(alpha = head(alpha, 1e4L), beta = head(beta, 1e4L))