diff --git a/backend/Cargo.toml b/backend/Cargo.toml
index 21279c2..c197214 100644
--- a/backend/Cargo.toml
+++ b/backend/Cargo.toml
@@ -12,4 +12,4 @@ opt-level = 3
 [profile.release]
 lto = true
 panic = "abort"
-debug = true
\ No newline at end of file
+# debug = true
\ No newline at end of file
diff --git a/backend/telemetry_core/tests/soak_tests.rs b/backend/telemetry_core/tests/soak_tests.rs
index 6b91a08..6b05248 100644
--- a/backend/telemetry_core/tests/soak_tests.rs
+++ b/backend/telemetry_core/tests/soak_tests.rs
@@ -16,38 +16,6 @@ sudo sysctl -w kern.ipc.maxsockbuf=16777216
 
 In general, if you run into issues, it may be better to run this on a linux
 box; MacOS seems to hit limits quicker in general.
-
-We can profile execution using cargo-flamegraph to learn more about how long
-we spend in each function.
-
-The main thing to do is install it:
-
-```sh
-cargo install flamegraph
-```
-
-Now, start processes running ourselves (so that we can use flamegraph on them):
-
-In one terminal (in the `backend` folder of this repo):
-
-```sh
-sudo cargo flamegraph -o telemetry_core.svg --bin telemetry_core
-```
-
-And in another:
-
-```sh
-sudo cargo flamegraph -o telemetry_shard.svg --bin telemetry_shard
-```
-
-And then, in a third terminal, we can run our general soak test against these:
-
-```sh
-SOAK_TEST_ARGS='--feeds 100 --nodes 100 --shards 4' \
-TELEMETRY_SUBMIT_HOSTS='127.0.0.1:8001' \
-TELEMETRY_FEED_HOST='127.0.0.1:8000' \
-cargo test -- soak_test --ignored --nocapture
-```
 */
 
 use futures::{ StreamExt };
@@ -67,12 +35,17 @@ use common::node_types::BlockHash;
 /// SOAK_TEST_ARGS='--feeds 10 --nodes 100 --shards 4' cargo test -- soak_test --ignored --nocapture
 /// ```
 ///
-/// You can also run this test against the pre-sharding actix binary like so:
+/// You can also run this test against the pre-sharding actix binary with something like this:
 /// ```sh
 /// TELEMETRY_BIN=~/old_telemetry_binary SOAK_TEST_ARGS='--feeds 100 --nodes 100 --shards 4' cargo test -- soak_test --ignored --nocapture
 /// ```
 ///
-/// Both will establish the same total number of connections and same the same messages.
+/// Or, you can run it against existing processes with something like this:
+/// ```sh
+/// TELEMETRY_SUBMIT_HOSTS='127.0.0.1:8001' TELEMETRY_FEED_HOST='127.0.0.1:8000' SOAK_TEST_ARGS='--feeds 100 --nodes 100 --shards 4' cargo test -- soak_test --ignored --nocapture
+/// ```
+///
+/// Each will establish the same total number of connections and send the same messages.
 #[ignore]
 #[tokio::test(flavor = "multi_thread", worker_threads = 4)]
 pub async fn soak_test() {
diff --git a/backend/test_utils/src/workspace/commands.rs b/backend/test_utils/src/workspace/commands.rs
index 595b2b2..df7f97a 100644
--- a/backend/test_utils/src/workspace/commands.rs
+++ b/backend/test_utils/src/workspace/commands.rs
@@ -21,7 +21,7 @@ fn telemetry_command(bin: &'static str, release_mode: bool) -> Result<Command, s
     let mut workspace_dir = try_find_workspace_dir()?;
     workspace_dir.push("Cargo.toml");
 
-    let mut cmd = Command::new("cargo").arg("flamegraph");
+    let mut cmd = Command::new("cargo").arg("run");
 
     // Release mode?
     if release_mode {