Confluent and IBM partner

Kafka setup, migration and support.
From a Confluent and IBM partner.

Get the Kafka you run today production-ready, or deploy Lascaris Streamhouse in your own environment.

How we work

Four steps. Start with any of them.

Most teams start with a short assessment and a written plan. Some only need engineers or support. Every offer has a fixed scope, agreed before we start.

Step 1 · Assess

Know what you run and what to change

We map your Apache Kafka, Confluent or Lascaris Streamhouse setup, then write down what to do next.

Step 2 · Build or migrate

Put the plan into production

We build it with your team in weekly sprints, with the scope fixed before each sprint starts.

Step 3 · Run

Someone to call when it breaks

Engineers inside your team, retainers with a 24-hour answer, and yearly support packages.

Step 4 · Enable

A team that can run it without us

Live cohorts and workshops for your architects, developers and operators.

White paper

From sensor to superintelligence.

How AI agents work on the Kafka you already run, and how you keep every one of their decisions under control. It covers why Kafka is the right foundation, how Lascaris checks each action against your policy before it runs, and where it runs in your own infrastructure.

  • The four questions to answer before your first agent goes live.
  • Sense, decide, check, record: one agent loop, step by step.
  • The architecture, the deployment options and how to start a pilot.

Step 1 · Assess

Start here.

Four ways to begin, all with a fixed scope. Each one ends with a written plan you can act on. Prices for every offer are on the pricing page.

3-5 days

Architecture sprint

A short sprint for Apache Kafka, Confluent or Lascaris Streamhouse. We map your topics, clients and costs.

What is included

  • Review of Confluent, Amazon MSK, WarpStream, Redpanda, Apache Kafka or Lascaris Streamhouse
  • Topic, client and cost mapping
  • Architecture and sizing for your Kubernetes and S3, if Lascaris Streamhouse is the target
  • Backup, recovery and security needs written into the plan

Who it is for

Platform engineering, SRE and data infrastructure teams who need a plan, not a slide deck.

What you get

A written migration or production-readiness plan, with the scope of the next sprints.

By the week

Design workshop

Your architects and ours at one whiteboard, until the design is written down.

What is included

  • Working sessions with your architects
  • From whiteboard to a written design
  • Run on site or remote

Who it is for

Teams starting or redesigning Apache Kafka, Confluent or Lascaris Streamhouse.

What you get

A written design your team can build from.

One-time or by the week

Performance tuning

For Apache Kafka, Confluent or Lascaris Streamhouse when the cluster is slow, expensive or close to its limits.

What is included

  • Tuning of the cluster and its producers and consumers
  • Capacity planning and cost review
  • As a one-time job, or as an engagement booked by the week

Who it is for

Platform engineering, SRE and data infrastructure teams running a cluster in production today.

What you get

A tuned cluster and a written record of what changed and why.

Fixed package

Consulting package

Assessment, two sprints and hand-over in one package, run in step with your own sprints.

What is included

  • Infrastructure assessment and planning
  • 2 sprints in sync with yours
  • Knowledge transfer and training (2 days)
  • Proactive solution ideas

Who it is for

Teams that want an assessment and the first changes done, not only advice.

What you get

A plan, the first work shipped, and a team that knows how it was built.

Step 2 · Build or migrate

Build and migrate.

Weekly sprints with a fixed scope agreed up front. Production keeps running, and nothing moves until you have a backup you control.

Implementation and migration sprints

We deploy and migrate in one-week sprints, for Apache Kafka, Confluent or Lascaris Streamhouse in your cloud, on-premise or air-gapped.

What is included

  • Deployment in your cloud, on-premise or air-gapped
  • Migration from Confluent, WarpStream, Redpanda or Amazon MSK
  • Backup setup, recovery runbooks and restore tests
  • Security implementation (RBAC, encryption, audit)
  • Capacity planning and cost review

Who it is for

Platform engineering, SRE and data infrastructure teams with a plan and a deadline.

What you get

A running platform, tested restores, and runbooks your team can use without us.

The low-risk switch path

Switching from Confluent, WarpStream or Amazon MSK is not an all or nothing move. You start with a backup you control and a written plan, then move workloads when you are ready.

How it works

  • First, an immutable backup of your current cluster with KafShield, on object storage you own
  • Your production cluster stays as it is. The backup only adds one read-only consumer
  • An architecture sprint maps your topics, clients and costs into a written migration plan
  • Then migration sprints, each with a fixed scope agreed up front

Who it is for

Teams on Confluent, WarpStream or Amazon MSK who want lower cost or more control, without a risky big-bang cut-over.

What you get

A backup you control before anything moves, and a migration done in steps you agree to. KafShield also works on its own, without Lascaris Streamhouse.

SAP, ERP and SCADA integration

Real-time data from enterprise systems into Apache Kafka or Lascaris Streamhouse, with custom Kafka Connect connectors.

What is included

  • SAP IDoc extraction and SAP object events into Kafka
  • ERP real-time data integration
  • SCADA data from smart grid and industrial sites
  • Event-driven architecture design
  • Custom Kafka Connect connectors

Who it is for

Manufacturing, energy and utilities, finance and retail teams with legacy ERP systems.

What you get

An architecture sprint, a discovery and proof of concept, then a production implementation with a fixed scope, and connectors built for your systems.

AI agents on Lascaris Streamhouse

Lascaris Streamhouse already ships governed AI agents, with KafClaw built in. We design and build production multi-agent systems on Kafka, with policy checks and an audit trail.

What is included

  • Multi-agent architecture design
  • Kafka-first communication patterns
  • Framework-agnostic agent abstractions
  • Production readiness reviews
  • Performance optimization and monitoring

Who it is for

AI platform teams and enterprise AI operations.

What you get

A one-week architecture sprint, implementation sprints of 2-4 weeks, then ongoing production optimization if you want it.

Apache Wayang in production

Production deployment, pipeline migration and architecture from the co-creators of Apache Wayang. Our team includes Apache Wayang PMC members and committers.

What is included

  • Wayang architecture design and review
  • Lascaris Streamhouse, Spark, Flink and JDBC integration
  • Cross-platform pipeline optimisation
  • Custom operator and platform extension
  • Production incident support and SLA retainers

Who it is for

Data platform teams moving off single-engine architectures, and teams running Apache Wayang in production.

What you get

An architecture sprint, implementation sprints, and an enterprise retainer for production.

Step 3 · Run

Engineers and support.

Some teams need hands on the keyboard for a few weeks. Others need someone on call for years. You can have either, or both.

Forward-deployed engineer

A senior Scalytics engineer who joins your team and does the work with you, not a consultant who writes a report and leaves.

What is included

  • A senior Scalytics engineer working inside your team
  • Works in your repos, your tickets and your stand-ups
  • Book by the week, stop when the work is done

Who it is for

Teams with more Apache Kafka, Confluent or Lascaris Streamhouse work than people, or a new team that needs senior experience from the first day.

What you get

Work shipped in your own repos, and a team that learned by building next to us.

Monthly

Standard retainer

An architect on call for Apache Kafka, Confluent or Lascaris Streamhouse, every month.

What is included

  • Monthly architecture office hours (4-8 hours)
  • Slack and email support, under 24h response
  • Production incident support
  • Architecture decision records
  • Quarterly roadmap reviews

Who it is for

Teams that run the platform themselves and want a second opinion and backup on hand.

Monthly

Premium retainer

Everything a production team needs when an outage costs real money.

What is included

  • Emergency on-call for production incidents
  • Slack and email support, under 24h response
  • Monthly architecture office hours (4-8 hours)
  • Quarterly roadmap reviews

Who it is for

Teams where Apache Kafka, Confluent or Lascaris Streamhouse carries business-critical traffic and incidents cannot wait.

Yearly packages

Lascaris Streamhouse support

Yearly support for Lascaris Streamhouse in your cloud or data center.

  • Onboarding configuration review (remote, 2h)
  • DORA Article 12 and NIS2 Annex II evidence pack
  • Quarterly DR test runbook
  • Updates and fixes for the supported release
  • Email support, 2-business-day response
See pricing

Lascaris Streamhouse Operated on IBM Cloud

We run Lascaris Streamhouse for you on IBM Cloud, as a support service.

  • Everything in Lascaris Streamhouse support
  • Kafka on S3 plus the immutable backup, operated by Scalytics on IBM Cloud
  • 1-business-day response
  • Monthly compliance reports, generated and signed
  • Multi-cluster configuration support
See pricing

DR Validation

An add-on to any support package, for teams that have to prove recovery to an auditor.

  • Annual recovery test run by Scalytics engineers
  • Segment integrity check and RTO against your SLA
  • Signed audit report for DORA Article 12 and NIS2 Annex II reviews
See pricing

Step 4 · Enable

Training and enablement.

Live cohorts, not video libraries. Architects, developers and operators train together. The cohorts are taught by Dr. Mirko Kämpf, a former Confluent Solution Architect, and delivered as a Confluent partner. Full programs are on the enablement page.

Live cohort · 3 weeks

Confluent Platform Onboarding

For teams running Confluent Platform on premises or on Kubernetes.

  • 15 live sessions of 90 minutes over three weeks
  • Hands-on labs and recordings
  • Foundation week for all roles, then developer, architect and operator tracks
  • Public cohort per seat, or a private cohort for up to 12 participants

Live cohort · 3-8 weeks

Confluent Cloud Training

The same structure, for teams running Confluent Cloud.

  • 15 live sessions of 90 minutes over three to eight weeks
  • Labs on RBAC, networking, governance and managed Flink
  • Architects, developers and operators in one program
  • Public cohort per seat, or a private cohort for up to 12 participants

2 days

Kafka Deep Dive

  • Architecture fundamentals
  • Performance tuning and optimization
  • Security and operations
  • Up to 15 participants
See pricing

Workshop

Apache Wayang Workshop

  • Cross-platform data processing
  • Implementation strategies
  • Cost optimization
  • Up to 15 participants
See pricing

By the day

Advanced training

  • Custom agenda for your team
  • On site or remote
See pricing

Why this team

Senior engineers, not middlemen.

⬡
Confluent partner and IBM partner
Scalytics is a Confluent partner and an IBM Implementation & Architecture Partner. We work with the Confluent and Apache Kafka you already run, and we can come in together with IBM.
Confluent partner IBM Implementation & Architecture Partner
⚡
Former Cloudera and Confluent engineers
Our team has worked at Allianz, Cloudera, Confluent, E.ON, Evariant and Google. Dr. Mirko Kämpf, who trains our cohorts, is a former Confluent Solution Architect.
Allianz Cloudera Confluent E.ON Evariant Google
🏢
Co-creators of Apache Wayang
Our team are co-creators of Apache Wayang, now an Apache Top-Level Project, with PMC members and committers on the team. We also write and maintain the open source, Apache 2.0 engine that Lascaris Streamhouse is built on.
PMC members Committers Apache 2.0
🔓
Your stack, your choice
We work on the Apache Kafka or Confluent you already run. Moving to Lascaris Streamhouse is an option, never a condition. KafShield, Kerovan and Helios work with Lascaris Streamhouse or with any Kafka.
Vendor-neutral Fixed-scope pricing Your stack
FAQ

Questions about our services

What do I pay for?

The work and the support, never a licence or a usage meter. Projects and sprints have a fixed scope and a fixed price agreed before we start. Engineers are booked by the week, retainers run by the month and support packages by the year. Every price is on the pricing page.

Do you work with the Apache Kafka or Confluent we already run?

Yes. We set up, migrate, tune and support Confluent Platform, Confluent Cloud, Amazon MSK and Apache Kafka, and we deploy and support Lascaris Streamhouse. KafShield backs up Confluent, Apache Kafka, Redpanda or Amazon MSK, and Kerovan and Helios run on Lascaris Streamhouse or on the Kafka you already have.

Do we have to move to Lascaris Streamhouse?

No. Lascaris Streamhouse is one of two paths. If the Apache Kafka or Confluent you run today is the right fit, we make it production-ready and support it. If you want to move later, we start with a backup and a written plan, so it is never an all or nothing move.

Can we start with the backup only?

Yes. You can buy KafShield on its own, without Lascaris Streamhouse, as a yearly Support package. It keeps an immutable backup of the Kafka you already run on object storage you own. Production is untouched, and you decide later whether to move any workloads.

How fast do you respond?

Lascaris Streamhouse support answers within 2 business days by email. Lascaris Streamhouse Operated on IBM Cloud answers within 1 business day. The Standard and Premium retainers answer within 24 hours on Slack or email, and the Premium retainer adds emergency on-call for production incidents.

Where should we start?

Most teams start with an architecture sprint of 3-5 days or a design workshop. Both end with a written plan. If you are not sure, tell us what you run today and what it costs you, and we come back with a written proposal.

Can you work with IBM?

Yes. Scalytics is an IBM Implementation & Architecture Partner, and we can come in together with IBM. We also run Lascaris Streamhouse for you on IBM Cloud with the Operated package.

Confluent and IBM partner

Get a written proposal.

Tell us what you run today, Apache Kafka, Confluent or Lascaris Streamhouse, and what it costs you. We come back with a written proposal: the services you need, the project scope and a fixed price.