diff --git a/examples/riverkfq/main.go b/examples/riverkfq/main.go
index 50eec76..eb03c02 100644
--- a/examples/riverkfq/main.go
+++ b/examples/riverkfq/main.go
@@ -38,22 +38,20 @@ func main() {
 	}
 
 	pool := createPool()
-
-	pq, err := riverkfq.NewPublishQueue(pool)
-	if err != nil {
-		panic(err)
-	}
-
 	producer := createProducer()
 
-	wq, err := riverkfq.NewRiverQueue(pool, producer)
+	pq, err := riverkfq.NewPublishQueue(
+		riverkfq.Pool(pool),
+		riverkfq.WithProducer(producer),
+		riverkfq.MaxWorkers(2),
+	)
 	if err != nil {
 		panic(err)
 	}
 
 	ctx, cancel := context.WithCancel(context.Background())
 
-	go wq.Run(ctx)
+	go pq.Run(ctx)
 
 	messages := generateMessages(topic, 10)
 
diff --git a/kfq/riverkfq/doc.go b/kfq/riverkfq/doc.go
new file mode 100644
index 0000000..a4e729e
--- /dev/null
+++ b/kfq/riverkfq/doc.go
@@ -0,0 +1,3 @@
+// Package riverkfq provides River based queues that offer
+// Kafka publishing guarantees.
+package riverkfq
diff --git a/kfq/riverkfq/options.go b/kfq/riverkfq/options.go
new file mode 100644
index 0000000..9e685d9
--- /dev/null
+++ b/kfq/riverkfq/options.go
@@ -0,0 +1,56 @@
+package riverkfq
+
+import (
+	"context"
+
+	"github.com/gojekfarm/xtools/xkafka"
+	"github.com/jackc/pgx/v5"
+	"github.com/jackc/pgx/v5/pgxpool"
+	"github.com/riverqueue/river"
+)
+
+// Option defines interface for configuring riverkfq
+type Option interface {
+	apply(*PublishQueue)
+}
+
+type optionFunc func(*PublishQueue)
+
+func (f optionFunc) apply(pq *PublishQueue) {
+	f(pq)
+}
+
+// Pool sets the database connection pool.
+func Pool(p *pgxpool.Pool) Option {
+	return optionFunc(func(pq *PublishQueue) {
+		pq.pool = p
+	})
+}
+
+// Producer is a Kafka producer
+type Producer interface {
+	Publish(ctx context.Context, msg *xkafka.Message) error
+}
+
+// WithProducer sets the Kafka producer.
+func WithProducer(p Producer) Option {
+	return optionFunc(func(pq *PublishQueue) {
+		pq.producer = p
+	})
+}
+
+// WithClient sets the river.Client.
+// Useful for integrating with existing river queues.
+func WithClient(c *river.Client[pgx.Tx]) Option {
+	return optionFunc(func(pq *PublishQueue) {
+		pq.queue = c
+	})
+}
+
+// MaxWorkers sets the maximum number of workers.
+// Default is 1.
+type MaxWorkers int
+
+func (m MaxWorkers) apply(pq *PublishQueue) {
+	pq.maxWorkers = int(m)
+}
diff --git a/kfq/riverkfq/river.go b/kfq/riverkfq/publish.go
similarity index 51%
rename from kfq/riverkfq/river.go
rename to kfq/riverkfq/publish.go
index 3869da1..0c56df2 100644
--- a/kfq/riverkfq/river.go
+++ b/kfq/riverkfq/publish.go
@@ -10,56 +10,47 @@ import (
 	"github.com/riverqueue/river/riverdriver/riverpgxv5"
 )
 
-// Producer is a Kafka producer
-type Producer interface {
-	Publish(ctx context.Context, msg *xkafka.Message) error
-}
+// PublishQueue is a thin wrapper around river.Client
+// that enqueues messages are asynchronously published to Kafka.
+type PublishQueue struct {
+	queue    *river.Client[pgx.Tx]
+	pool     *pgxpool.Pool
+	producer Producer
 
-// RiverQueue guarantees that messages are published
-// to Kafka.
-type RiverQueue struct {
-	queue *river.Client[pgx.Tx]
+	// config
+	maxWorkers int
 }
 
-// NewPublishQueue creates a new RiverQueue which can only
-// queue messages.
-func NewPublishQueue(pool *pgxpool.Pool) (*RiverQueue, error) {
-	client, err := river.NewClient(
-		riverpgxv5.New(pool),
-		&river.Config{},
-	)
-	if err != nil {
-		return nil, err
+// NewPublishQueue creates a new PublishQueue.
+func NewPublishQueue(opts ...Option) (*PublishQueue, error) {
+	pq := &PublishQueue{}
+
+	for _, opt := range opts {
+		opt.apply(pq)
 	}
 
-	return &RiverQueue{
-		queue: client,
-	}, nil
-}
+	if pq.queue != nil {
+		workers := river.NewWorkers()
+		river.AddWorker(workers, NewPublishWorker(pq.producer))
+
+		client, err := river.NewClient(riverpgxv5.New(pq.pool), &river.Config{
+			Queues: map[string]river.QueueConfig{
+				river.QueueDefault: {MaxWorkers: pq.maxWorkers},
+			},
+			Workers: workers,
+		})
+		if err != nil {
+			return nil, err
+		}
 
-// NewRiverQueue creates a new RiverQueue which can publish
-// messages to Kafka.
-func NewRiverQueue(pool *pgxpool.Pool, producer Producer) (*RiverQueue, error) {
-	workers := river.NewWorkers()
-	river.AddWorker(workers, &PublishWorker{producer: producer})
-
-	client, err := river.NewClient(riverpgxv5.New(pool), &river.Config{
-		Queues: map[string]river.QueueConfig{
-			river.QueueDefault: {MaxWorkers: 3},
-		},
-		Workers: workers,
-	})
-	if err != nil {
-		return nil, err
+		pq.queue = client
 	}
 
-	return &RiverQueue{
-		queue: client,
-	}, nil
+	return pq, nil
 }
 
 // Add enqueues one or more messages.
-func (q *RiverQueue) Add(ctx context.Context, msgs ...*xkafka.Message) error {
+func (q *PublishQueue) Add(ctx context.Context, msgs ...*xkafka.Message) error {
 	args := make([]river.InsertManyParams, len(msgs))
 
 	for i, msg := range msgs {
@@ -74,7 +65,7 @@ func (q *RiverQueue) Add(ctx context.Context, msgs ...*xkafka.Message) error {
 }
 
 // AddTx enqueues one or more messages in a transaction.
-func (q *RiverQueue) AddTx(ctx context.Context, tx pgx.Tx, msgs ...*xkafka.Message) error {
+func (q *PublishQueue) AddTx(ctx context.Context, tx pgx.Tx, msgs ...*xkafka.Message) error {
 	args := make([]river.InsertManyParams, len(msgs))
 
 	for i, msg := range msgs {
@@ -89,7 +80,7 @@ func (q *RiverQueue) AddTx(ctx context.Context, tx pgx.Tx, msgs ...*xkafka.Messa
 }
 
 // Run starts the queue and begins publishing messages to Kafka.
-func (q *RiverQueue) Run(ctx context.Context) error {
+func (q *PublishQueue) Run(ctx context.Context) error {
 	if err := q.queue.Start(ctx); err != nil {
 		return err
 	}
@@ -103,25 +94,28 @@ func (q *RiverQueue) Run(ctx context.Context) error {
 	return nil
 }
 
+// PublishArgs is the job payload for publishing messages to Kafka.
 type PublishArgs struct {
 	Msg *xkafka.Message `json:"msg"`
 }
 
+// Kind returns the kind of the job.
 func (args PublishArgs) Kind() string {
 	return "publish_xkafka_message"
 }
 
+// PublishWorker is a worker that publishes messages to Kafka.
 type PublishWorker struct {
 	river.WorkerDefaults[PublishArgs]
 	producer Producer
 }
 
+// NewPublishWorker creates a new PublishWorker.
 func NewPublishWorker(producer Producer) *PublishWorker {
-	return &PublishWorker{
-		producer: producer,
-	}
+	return &PublishWorker{producer: producer}
 }
 
+// Work executes the job.
 func (w *PublishWorker) Work(ctx context.Context, job *river.Job[PublishArgs]) error {
 	return w.producer.Publish(ctx, job.Args.Msg)
 }
diff --git a/kfq/riverkfq/river_test.go b/kfq/riverkfq/publish_test.go
similarity index 100%
rename from kfq/riverkfq/river_test.go
rename to kfq/riverkfq/publish_test.go