diff --git a/.cargo/config.toml b/.cargo/config.toml
index 40f5b553..7ec672b9 100644
--- a/.cargo/config.toml
+++ b/.cargo/config.toml
@@ -1,3 +1,4 @@
 [alias]
 poseidon-example = ["run", "--example", "poseidon", "--release", "--", "--json"]
 trivial-example = ["run", "--example", "trivial", "--release", "--", "--json"]
+cli-example = ["run", "--example", "cli", "--release", "--"]
diff --git a/Cargo.toml b/Cargo.toml
index cdc3e67f..1522addb 100644
--- a/Cargo.toml
+++ b/Cargo.toml
@@ -27,14 +27,15 @@ thiserror = "1.0.48"
 tracing = { version = "0.1.40", features = ["attributes"] }
 
 [dev-dependencies]
-prettytable-rs = "0.10.0"
-tracing-test = "0.2.4"
-halo2_gadgets = { git = "https://github.com/snarkify/halo2", branch = "snarkify/dev", features = ["unstable"] }
 bincode = "1.3"
+clap = { version = "4.5.4", features = ["derive"] }
+criterion = "0.5.1"
+halo2_gadgets = { git = "https://github.com/snarkify/halo2", branch = "snarkify/dev", features = ["unstable"] }
+maplit = "1.0.2"
+prettytable-rs = "0.10.0"
 tempfile = "3.9.0"
 tracing-subscriber = { version = "0.3.18", features = ["json"] }
-maplit = "1.0.2"
-criterion = "0.5.1"
+tracing-test = "0.2.4"
 
 [dev-dependencies.cargo-husky]
 version = "1"
diff --git a/README.md b/README.md
index 6f0559c5..5b22d5b6 100644
--- a/README.md
+++ b/README.md
@@ -146,6 +146,9 @@ This code will run `fold_step_count` of folding steps, and also check the proof
 For runnable examples, please check [examples](examples) folder.
 
 ```bash
+# Alias to run IVC with parameterization via cli arguments
+cargo cli-example --help
+
 # Alias for run the IVC for trivial `StepCircuit` (just returns its input unchanged)
 cargo trivial-example
 
diff --git a/examples/cli.rs b/examples/cli.rs
new file mode 100644
index 00000000..bb26ee7e
--- /dev/null
+++ b/examples/cli.rs
@@ -0,0 +1,207 @@
+use std::{array, num::NonZeroUsize};
+
+use clap::{Parser, ValueEnum};
+
+#[allow(dead_code)]
+mod poseidon;
+
+use ff::Field;
+use poseidon::poseidon_step_circuit::TestPoseidonCircuit;
+use sirius::{
+    ivc::{step_circuit::trivial, CircuitPublicParamsInput, PublicParams, StepCircuit, IVC},
+    poseidon::ROPair,
+};
+use tracing::*;
+use tracing_subscriber::{filter::LevelFilter, fmt::format::FmtSpan, EnvFilter};
+
+#[derive(Parser, Debug)]
+#[command(version, about, long_about = None)]
+struct Args {
+    #[arg(value_enum, default_value_t = Circuits::Poseidon)       ]
+    primary_circuit: Circuits,
+    #[arg(long, default_value_t = 17)]
+    primary_circuit_k_table_size: u32,
+    #[arg(long, default_value_t = 21)]
+    primary_commitment_key_size: usize,
+    #[arg(long, default_value_t = 1)]
+    primary_repeat_count: usize,
+    #[arg(long, default_value_t = 10)]
+    primary_r_f: usize,
+    #[arg(long, default_value_t = 10)]
+    primary_r_p: usize,
+    #[arg(value_enum, default_value_t = Circuits::Trivial)        ]
+    secondary_circuit: Circuits,
+    #[arg(long, default_value_t = 17)]
+    secondary_circuit_k_table_size: u32,
+    #[arg(long, default_value_t = 21)]
+    secondary_commitment_key_size: usize,
+    #[arg(long, default_value_t = 1)]
+    secondary_repeat_count: usize,
+    #[arg(long, default_value_t = 10)]
+    secondary_r_f: usize,
+    #[arg(long, default_value_t = 10)]
+    secondary_r_p: usize,
+    #[arg(long, default_value_t = NonZeroUsize::new(32).unwrap()) ]
+    limb_width: NonZeroUsize,
+    #[arg(long, default_value_t = NonZeroUsize::new(10).unwrap()) ]
+    limbs_count: NonZeroUsize,
+    #[arg(long, default_value_t = false)]
+    debug_mode: bool,
+    #[arg(long, default_value_t = NonZeroUsize::new(1).unwrap())  ]
+    fold_step_count: NonZeroUsize,
+    #[arg(long, default_value_t = false)]
+    json_logs: bool,
+}
+
+#[derive(Copy, Clone, PartialEq, Eq, PartialOrd, Ord, ValueEnum, Debug)]
+enum Circuits {
+    Poseidon,
+    Trivial,
+}
+
+use halo2curves::{bn256, grumpkin};
+
+use bn256::G1 as C1;
+use grumpkin::G1 as C2;
+
+const MAIN_GATE_SIZE: usize = 5;
+const RATE: usize = 4;
+
+type RandomOracle = sirius::poseidon::PoseidonRO<MAIN_GATE_SIZE, RATE>;
+type RandomOracleConstant<F> = <RandomOracle as ROPair<F>>::Args;
+
+type C1Affine = <C1 as halo2curves::group::prime::PrimeCurve>::Affine;
+type C1Scalar = <C1 as halo2curves::group::Group>::Scalar;
+
+type C2Affine = <C2 as halo2curves::group::prime::PrimeCurve>::Affine;
+type C2Scalar = <C2 as halo2curves::group::Group>::Scalar;
+
+fn fold(
+    args: &Args,
+    primary: impl StepCircuit<1, C1Scalar>,
+    secondary: impl StepCircuit<1, C2Scalar>,
+) {
+    let primary_commitment_key = poseidon::get_or_create_commitment_key::<C1Affine>(
+        args.primary_commitment_key_size,
+        "bn256",
+    )
+    .expect("Failed to get primary key");
+    let secondary_commitment_key = poseidon::get_or_create_commitment_key::<C2Affine>(
+        args.secondary_commitment_key_size,
+        "grumpkin",
+    )
+    .expect("Failed to get secondary key");
+
+    // Specifications for random oracle used as part of the IVC algorithm
+    let primary_spec = RandomOracleConstant::<C1Scalar>::new(args.primary_r_f, args.primary_r_p);
+    let secondary_spec =
+        RandomOracleConstant::<C2Scalar>::new(args.secondary_r_f, args.secondary_r_p);
+
+    let pp = PublicParams::<
+        '_,
+        1,
+        1,
+        MAIN_GATE_SIZE,
+        C1Affine,
+        C2Affine,
+        _,
+        _,
+        RandomOracle,
+        RandomOracle,
+    >::new(
+        CircuitPublicParamsInput::new(
+            args.primary_circuit_k_table_size,
+            &primary_commitment_key,
+            primary_spec,
+            &primary,
+        ),
+        CircuitPublicParamsInput::new(
+            args.secondary_circuit_k_table_size,
+            &secondary_commitment_key,
+            secondary_spec,
+            &secondary,
+        ),
+        args.limb_width,
+        args.limbs_count,
+    )
+    .unwrap();
+
+    let mut rnd = rand::thread_rng();
+    let primary_input = array::from_fn(|_| C1Scalar::random(&mut rnd));
+    let secondary_input = array::from_fn(|_| C2Scalar::random(&mut rnd));
+
+    if args.debug_mode {
+        IVC::fold_with_debug_mode(
+            &pp,
+            primary,
+            primary_input,
+            secondary,
+            secondary_input,
+            args.fold_step_count,
+        )
+        .unwrap();
+    } else {
+        IVC::fold(
+            &pp,
+            primary,
+            primary_input,
+            secondary,
+            secondary_input,
+            args.fold_step_count,
+        )
+        .unwrap();
+    }
+}
+
+fn main() {
+    let args = Args::parse();
+
+    let builder = tracing_subscriber::fmt()
+        // Adds events to track the entry and exit of the span, which are used to build
+        // time-profiling
+        .with_span_events(FmtSpan::ENTER | FmtSpan::CLOSE)
+        // Changes the default level to INFO
+        .with_env_filter(
+            EnvFilter::builder()
+                .with_default_directive(LevelFilter::INFO.into())
+                .from_env_lossy(),
+        );
+
+    // Structured logs are needed for time-profiling, while for simple run regular logs are
+    // more convenient.
+    //
+    // So this expr keeps track of the --json argument for turn-on json-logs
+    if args.json_logs {
+        builder.json().init();
+    } else {
+        builder.init();
+    }
+
+    // To osterize the total execution time of the example
+    let _span = info_span!("poseidon_example").entered();
+
+    // Such a redundant call design due to the fact that they are different function types for the
+    // compiler due to generics
+    match (args.primary_circuit, args.secondary_circuit) {
+        (Circuits::Poseidon, Circuits::Trivial) => fold(
+            &args,
+            TestPoseidonCircuit::new(args.primary_repeat_count),
+            trivial::Circuit::default(),
+        ),
+        (Circuits::Poseidon, Circuits::Poseidon) => fold(
+            &args,
+            TestPoseidonCircuit::new(args.primary_repeat_count),
+            TestPoseidonCircuit::new(args.secondary_repeat_count),
+        ),
+        (Circuits::Trivial, Circuits::Poseidon) => fold(
+            &args,
+            trivial::Circuit::default(),
+            TestPoseidonCircuit::new(args.secondary_repeat_count),
+        ),
+        (Circuits::Trivial, Circuits::Trivial) => fold(
+            &args,
+            trivial::Circuit::default(),
+            trivial::Circuit::default(),
+        ),
+    }
+}
diff --git a/examples/poseidon/main.rs b/examples/poseidon.rs
similarity index 74%
rename from examples/poseidon/main.rs
rename to examples/poseidon.rs
index 8ade6a12..f2ff97a9 100644
--- a/examples/poseidon/main.rs
+++ b/examples/poseidon.rs
@@ -1,5 +1,5 @@
 /// This module represents an implementation of `StepCircuit` based on the poseidon chip
-mod poseidon_step_circuit {
+pub mod poseidon_step_circuit {
     use std::marker::PhantomData;
 
     use ff::{FromUniformBytes, PrimeFieldBits};
@@ -13,6 +13,7 @@ mod poseidon_step_circuit {
         main_gate::{MainGate, MainGateConfig, RegionCtx, WrapValue},
         poseidon::{poseidon_circuit::PoseidonChip, Spec},
     };
+    use tracing::*;
 
     /// Input and output size for `StepCircuit` within each step
     pub const ARITY: usize = 1;
@@ -21,7 +22,7 @@ mod poseidon_step_circuit {
     const POSEIDON_PERMUTATION_WIDTH: usize = 3;
     const POSEIDON_RATE: usize = POSEIDON_PERMUTATION_WIDTH - 1;
 
-    type CircuitPoseidonSpec<F> = Spec<F, POSEIDON_PERMUTATION_WIDTH, POSEIDON_RATE>;
+    pub type CircuitPoseidonSpec<F> = Spec<F, POSEIDON_PERMUTATION_WIDTH, POSEIDON_RATE>;
 
     const R_F1: usize = 4;
     const R_P1: usize = 3;
@@ -31,11 +32,30 @@ mod poseidon_step_circuit {
         pconfig: MainGateConfig<POSEIDON_PERMUTATION_WIDTH>,
     }
 
-    #[derive(Default, Debug)]
+    #[derive(Debug)]
     pub struct TestPoseidonCircuit<F: PrimeFieldBits> {
+        repeat_count: usize,
         _p: PhantomData<F>,
     }
 
+    impl<F: PrimeFieldBits> Default for TestPoseidonCircuit<F> {
+        fn default() -> Self {
+            Self {
+                repeat_count: 1,
+                _p: Default::default(),
+            }
+        }
+    }
+
+    impl<F: PrimeFieldBits> TestPoseidonCircuit<F> {
+        pub fn new(repeat_count: usize) -> Self {
+            Self {
+                repeat_count,
+                _p: Default::default(),
+            }
+        }
+    }
+
     impl<F: PrimeFieldBits + FromUniformBytes<64>> StepCircuit<ARITY, F> for TestPoseidonCircuit<F> {
         type Config = TestPoseidonCircuitConfig;
 
@@ -51,19 +71,50 @@ mod poseidon_step_circuit {
             z_in: &[AssignedCell<F, F>; ARITY],
         ) -> Result<[AssignedCell<F, F>; ARITY], SynthesisError> {
             let spec = CircuitPoseidonSpec::<F>::new(R_F1, R_P1);
-            let mut pchip = PoseidonChip::new(config.pconfig, spec);
-            let input = z_in.iter().map(|x| x.into()).collect::<Vec<WrapValue<F>>>();
-            pchip.update(&input);
-            let output = layouter
+
+            layouter
                 .assign_region(
                     || "poseidon hash",
-                    |region| {
+                    move |region| {
+                        let mut z_i = z_in.clone();
                         let ctx = &mut RegionCtx::new(region, 0);
-                        pchip.squeeze(ctx)
+
+                        for step in 0..=self.repeat_count {
+                            let mut pchip = PoseidonChip::new(config.pconfig.clone(), spec.clone());
+
+                            pchip.update(
+                                &z_i.iter()
+                                    .cloned()
+                                    .map(WrapValue::Assigned)
+                                    .collect::<Vec<WrapValue<F>>>(),
+                            );
+
+                            info!(
+                                "offset for {} hash repeat count is {} (log2 = {})",
+                                step,
+                                ctx.offset(),
+                                (ctx.offset() as f64).log2()
+                            );
+
+                            z_i = [pchip.squeeze(ctx).inspect_err(|err| {
+                                error!("at step {step}: {err:?}");
+                            })?];
+                        }
+
+                        info!(
+                            "total offset for {} hash repeat count is {} (log2 = {})",
+                            self.repeat_count,
+                            ctx.offset(),
+                            (ctx.offset() as f64).log2()
+                        );
+
+                        Ok(z_i)
                     },
                 )
-                .map_err(SynthesisError::Halo2)?;
-            Ok([output])
+                .map_err(|err| {
+                    error!("while synth {err:?}");
+                    SynthesisError::Halo2(err)
+                })
         }
     }
 }
@@ -117,7 +168,7 @@ type C2Scalar = <C2 as halo2curves::group::Group>::Scalar;
 
 /// Either takes the key from [`CACHE_FOLDER`] or generates a new one and puts it in it
 #[instrument]
-fn get_or_create_commitment_key<C: CurveAffine>(
+pub fn get_or_create_commitment_key<C: CurveAffine>(
     k: usize,
     label: &'static str,
 ) -> io::Result<CommitmentKey<C>> {