diff --git a/lib/pls/include/pls/dataflow/internal/function_node_impl.h b/lib/pls/include/pls/dataflow/internal/function_node_impl.h
new file mode 100644
index 0000000..3f2d3b0
--- /dev/null
+++ b/lib/pls/include/pls/dataflow/internal/function_node_impl.h
@@ -0,0 +1,85 @@
+
+#ifndef PLS_DATAFLOW_INTERNAL_FUNCTION_NODE_IMPL_H_
+#define PLS_DATAFLOW_INTERNAL_FUNCTION_NODE_IMPL_H_
+
+namespace pls {
+namespace dataflow {
+namespace internal {
+
+template<typename I0, typename ...I, typename O0, typename ...O, typename F>
+template<int POS, typename T>
+void function_node<pls::dataflow::inputs<I0, I...>, pls::dataflow::outputs<O0, O...>, F>::
+token_pushed(token<T> token) {
+  auto current_memory = get_invocation<invocation_memory>(token.invocation());
+
+  std::get<POS>(current_memory->input_buffer_) = token;
+  auto remaining_inputs = --(current_memory->inputs_missing_);
+  if (remaining_inputs == 0) {
+    execute_function(current_memory, token.invocation());
+    current_memory->inputs_missing_ = num_in_ports;
+  }
+}
+
+//////////////////////////////////////////////////////////////////
+// Helpers for actually calling the work lambda
+//////////////////////////////////////////////////////////////////
+template<typename I0, typename ...I, typename O0, typename ...O, typename F>
+void function_node<pls::dataflow::inputs<I0, I...>, pls::dataflow::outputs<O0, O...>, F>::
+execute_function(invocation_memory *invocation_memory, invocation_info invocation_info) {
+  auto lambda = [&]() {
+    input_tuple &inputs = invocation_memory->input_buffer_;
+    output_tuple outputs{};
+
+    execute_function_internal(inputs, typename sequence_gen<1 + sizeof...(I)>::type(),
+                              outputs, typename sequence_gen<1 + sizeof...(O)>::type(), invocation_info);
+  };
+  // TODO: maybe replace this with 'continuation' style invocation
+  pls::scheduler::spawn_child_and_wait<pls::lambda_task_by_reference<decltype(lambda)>>(lambda);
+}
+
+template<typename I0, typename ...I, typename O0, typename ...O, typename F>
+template<int N, typename ...OT>
+struct function_node<pls::dataflow::inputs<I0, I...>, pls::dataflow::outputs<O0, O...>, F>::
+propagate_output {
+  propagate_output(multi_out_port_type &, output_tuple &) {}
+  void propagate() {}
+};
+template<typename I0, typename ...I, typename O0, typename ...O, typename F>
+template<int N, typename OT1, typename ...OT>
+struct function_node<pls::dataflow::inputs<I0, I...>, pls::dataflow::outputs<O0, O...>, F>::
+propagate_output<N, OT1, OT...> {
+  multi_out_port_type &out_port_;
+  output_tuple &output_tuple_;
+
+  propagate_output(multi_out_port_type &out_port, output_tuple &output_tuple) : out_port_{out_port},
+                                                                                output_tuple_{output_tuple} {}
+
+  void propagate() {
+    out_port_.template get<N>().push_token(std::get<N>(output_tuple_));
+    propagate_output<N + 1, OT...>{out_port_, output_tuple_}.propagate();
+  }
+};
+
+template<typename I0, typename ...I, typename O0, typename ...O, typename F>
+template<typename T>
+void function_node<pls::dataflow::inputs<I0, I...>, pls::dataflow::outputs<O0, O...>, F>::
+set_invocation_info(token<T> &token, invocation_info invocation_info) {
+  token.set_invocation(invocation_info);
+}
+
+template<typename I0, typename ...I, typename O0, typename ...O, typename F>
+template<int ...IS, int ...OS>
+void function_node<pls::dataflow::inputs<I0, I...>, pls::dataflow::outputs<O0, O...>, F>::
+execute_function_internal(input_tuple &inputs, sequence<IS...>,
+                          output_tuple &outputs, sequence<OS...>,
+                          invocation_info invocation_info) {
+  set_invocation_info(std::get<OS>(outputs)..., invocation_info);
+  function_(std::get<IS>(inputs).value()..., std::get<OS>(outputs).value()...);
+  propagate_output<0, O0, O...>{out_port_, outputs}.propagate();
+}
+
+}
+}
+}
+
+#endif //PLS_DATAFLOW_INTERNAL_FUNCTION_NODE_IMPL_H_
diff --git a/lib/pls/include/pls/dataflow/internal/graph_impl.h b/lib/pls/include/pls/dataflow/internal/graph_impl.h
new file mode 100644
index 0000000..ccdc04c
--- /dev/null
+++ b/lib/pls/include/pls/dataflow/internal/graph_impl.h
@@ -0,0 +1,131 @@
+
+#ifndef PLS_DATAFLOW_INTERNAL_GRAPH_IMPL_H_
+#define PLS_DATAFLOW_INTERNAL_GRAPH_IMPL_H_
+
+namespace pls {
+namespace dataflow {
+namespace internal {
+
+template<typename I0, typename ...I, typename O0, typename ...O>
+template<int POS, typename T>
+void graph<pls::dataflow::inputs<I0, I...>, pls::dataflow::outputs<O0, O...>>::
+token_pushed(token<T> token) {
+  auto invocation = get_invocation<invocation_memory>(token.invocation());
+  std::get<POS>(*invocation->output_buffer_) = token.value();
+}
+
+template<typename I0, typename ...I, typename O0, typename ...O>
+void graph<pls::dataflow::inputs<I0, I...>, pls::dataflow::outputs<O0, O...>>::
+build() {
+  PLS_ASSERT(build_state_ == build_state::fresh, "Must only build a dataflow graph once!")
+  PLS_ASSERT(is_fully_connected(), "Must fully connect all inputs/outputs inside a dataflow graph!")
+
+  build_state_ = build_state::building;
+  node_list_start_ = node_list_current_ = this;
+  memory_index_ = 0;
+  num_nodes_ = 1;
+  for (int i = 0; i < num_in_ports; i++) {
+    build_recursive(inputs_.next_node_at(i));
+  }
+  build_state_ = build_state::built;
+}
+
+template<typename I0, typename ...I, typename O0, typename ...O>
+void graph<pls::dataflow::inputs<I0, I...>, pls::dataflow::outputs<O0, O...>>::
+build_recursive(node *node) {
+  if (node->build_state_ != build_state::fresh) {
+    return; // Already visited
+  }
+  node->build_state_ = build_state::building;
+  PLS_ASSERT(node->is_fully_connected(), "Must fully connect dataflow graph nodes!")
+
+  add_node(node);
+  for (int i = 0; i < node->num_successors(); i++) {
+    build_recursive(node->successor_at(i));
+  }
+
+  node->build_state_ = build_state::built;
+}
+
+template<typename I0, typename ...I, typename O0, typename ...O>
+void graph<pls::dataflow::inputs<I0, I...>, pls::dataflow::outputs<O0, O...>>::
+add_node(node *new_node) {
+  new_node->memory_index_ = num_nodes_++;
+
+  if (node_list_current_ == nullptr) {
+    node_list_current_ = new_node;
+    node_list_start_ = new_node;
+  } else {
+    node_list_current_->direct_successor_ = new_node;
+    node_list_current_ = new_node;
+  }
+}
+
+template<typename I0, typename ...I, typename O0, typename ...O>
+template<int N, typename ...IT>
+struct graph<pls::dataflow::inputs<I0, I...>, pls::dataflow::outputs<O0, O...>>::
+feed_inputs {
+  feed_inputs(inputs_type &, value_input_tuple &, invocation_info &) {}
+  void run() {}
+};
+
+template<typename I0, typename ...I, typename O0, typename ...O>
+template<int N, typename IT1, typename ...IT>
+struct graph<pls::dataflow::inputs<I0, I...>, pls::dataflow::outputs<O0, O...>>::
+feed_inputs<N, IT1, IT...> {
+  inputs_type &inputs_;
+  value_input_tuple &input_values_;
+  invocation_info &invocation_;
+
+  feed_inputs(inputs_type &inputs,
+              value_input_tuple &input_values,
+              invocation_info &invocation) : inputs_{inputs},
+                                             input_values_{input_values},
+                                             invocation_{invocation} {}
+
+  void run() {
+    inputs_.template get<N>().push_token(token<IT1>{std::get<N>(input_values_), invocation_});
+    feed_inputs<N + 1, IT...>{inputs_, input_values_, invocation_}.run();
+  }
+};
+
+template<typename I0, typename ...I, typename O0, typename ...O>
+class graph<pls::dataflow::inputs<I0, I...>, pls::dataflow::outputs<O0, O...>>::run_graph_task : public pls::task {
+  graph *self_;
+  value_input_tuple input_;
+  value_output_tuple *output_;
+
+  // Buffers for actual execution
+  invocation_info invocation_;
+
+ public:
+  run_graph_task(graph *self, value_input_tuple &input, value_output_tuple *output) : self_{self},
+                                                                                      input_{input},
+                                                                                      output_{output},
+                                                                                      invocation_{nullptr} {
+    void **buffers;
+    buffers = reinterpret_cast<void **>(allocate_memory(self_->num_nodes_ * sizeof(void *)));
+
+    node *iterator = self_->node_list_start_;
+    for (int i = 0; i < self_->num_nodes_; i++) {
+      auto required_size = iterator->instance_buffer_size();
+      buffers[i] = allocate_memory(required_size);
+      iterator->init_instance_buffer(buffers[i]);
+
+      iterator = iterator->direct_successor_;
+    }
+    invocation_ = invocation_info{buffers};
+
+    self_->get_invocation<invocation_memory>(invocation_)->output_buffer_ = output;
+  }
+
+  void execute_internal() override {
+    feed_inputs<0, I0, I...>{self_->inputs_, input_, invocation_}.run();
+  }
+};
+
+}
+}
+}
+
+#endif //PLS_DATAFLOW_INTERNAL_GRAPH_IMPL_H_