Talking Heads sang "Once in a Lifetime" in 1980. Its central image - water flowing underground, endlessly, whether anyone's watching or not - was about sleepwalking through your own life while forces beneath the surface keep moving without you. It's also, unintentionally, a fair description of how enterprise data behaves: constantly flowing beneath the business, mostly unseen, doing its job whether a human ever looks at it or not. What's different isn't the flow. It's who's drinking from it. Qlik, SAP, and a cluster of vendors at API World have all recently opened the tap to a new consumer - AI agents -via a shared standard called MCP (Model Context Protocol). Same water. Different drinker.
For years, new Qlik Cloud capabilities have arrived the same way: finished. That sounds good until you notice what it means. We only hear what the broader user base thinks once something is generally available, which is the exact moment when changing it is most expensive.
Ask your coding agent a real question about your Qlik tenant and get a factual answer in real time. Qlik's MCP server now ships three new lookup tools purpose-built for declarative pipeline development, so Claude Code, Copilot, and Codex can check spaces, connections, and table structures against your live environment instead of guessing. No more tabbing out to the Qlik Cloud web app to hunt down a connection name or table ID, just faster, more accurate pipelines from the first prompt.
Qlik Talend Unlocked is a monthly series where engineers who have shipped this into production show you how. Each session takes on a real part of the modern Talend lifecycle, from CI/CD job promotion and test automation to migration, open lakehouse architectures, and real time integration. Practical approaches you can take straight back to your own environment, with results teams can measure.
Episode one is on demand now, episode two is open for registration. Register once and you are set for the whole series.
Something significant has shifted in the SAP ecosystem. SAP first prohibited ODP-RFC access for third-party tools back in February 2024. For over two years, that rule existed on paper without technical enforcement. In April 2026, that changed: SAP announced active enforcement was coming, and in June 2026, it arrived. Here is your plain-language explanation of what it means for Qlik customers, and where Qlik goes from here.
Private Preview · Google Cloud
Qlik Open Lakehouse comes to Google Cloud
Sign up for the Private Preview and get hands-on with real-time Iceberg tables in your own Google Cloud project — no lock-in, no re-architecting.
Sign up for the Private Preview!
If you're running analytics on Google Cloud, you already know the tension. BigQuery is fast and familiar, but the more data you pour into it, the more your ingestion and storage bill climbs, and the more your data ends up shaped around one engine. You want the performance without the lock-in, you want your data to stay yours, and you want to free up some budget for more valuable workloads.
That's exactly the problem Qlik Open Lakehouse solves — and we're getting ready to launch a Private Preview on Google Cloud this September that you can be part of.
Qlik Open Lakehouse gives you a fully managed way to ingest real-time data straight into Apache Iceberg tables that live in your own Google Cloud Storage buckets. No proprietary formats, no copies, no re-architecting your data layer. Query those same tables from BigQuery, Trino, or any Iceberg-compatible engine, and pay a fraction of what you'd normally spend moving and storing that data.
If you've followed our work on AWS, this will feel familiar — and that's the point. We're porting a proven architecture to Google Cloud, not building a science experiment.
How it works: a walk through the architecture
Open Lakehouse builds the bronze layer of your medallion architecture, ingesting and landing real-time data as optimized Apache Iceberg tables, with the heavy lifting executed in your own Google Cloud project. From there, push-down transformations in Qlik Talend® Cloud pipelines create your silver and gold layers directly on BigQuery. Here's the path your data takes, stage by stage.
1 · Ingestion at the edge
Your sources land through two paths. For change data capture and bulk loads from databases, SAP, and mainframe systems, the Qlik Data Movement Gateway sits in your managed environment and handles CDC and bulk ingestion. For streaming and event data, you connect Kafka topics or Google Cloud Storage directly. Both routes feed the landing stage without your data ever leaving your control. Later in the Private Preview, we'll also add support for Qlik Replicate as a source, so if you're already a Replicate customer, you can onboard data from your existing CDC sources into Iceberg without migrating a single pipeline or switching tools.
2 · Compute in your project
Two components run on Google Compute Engine inside your environment. The Network Integration Agent runs on a standard Compute Engine VM and brokers secure connectivity between the Qlik Talend Cloud tenant and your resources. The Open Lakehouse Cluster — the engine that writes and optimizes your Iceberg tables — runs on a Compute Engine managed instance group built on Spot VMs. That Spot-based design is deliberate: it's where a large part of the cost savings comes from, because you're running the ingestion and optimization workload on Google Cloud's most cost-efficient compute instead of always-on, premium-priced infrastructure.
3 · Open storage you own
Everything lands in Google Cloud Storage, split across a landing bucket for incoming data, a config bucket for pipeline metadata, and table storage where your data comes to rest as Apache Iceberg tables. These buckets are in your account. The data is written in an open table format from the start, so there's never a moment where it's trapped in something only Qlik can read.
4 · A catalog any engine can use
Iceberg tables are registered in the Google Lakehouse runtime catalog, so your tables are discoverable and governable through Google Cloud's native catalog rather than a Qlik-proprietary one. That's what makes zero-copy querying real: BigQuery reads the same tables through ELT, and so can Trino or any other Iceberg-compatible engine, all pointing at one physical copy of the data.
5 · A managed bronze layer, ready to build on
The pipeline moves through managed stages — landing, storage, and mirror — to produce clean, governed Iceberg tables without you hand-coding each hop. You get Type 1 and Type 2 slowly changing dimensions or append-only patterns out of the box, so history and current-state tables are handled for you. We'll also add zero-copy mirroring to BigQuery during the Private Preview, so you can surface your Iceberg data straight in BigQuery. If you want to replace an existing native table, you can, since downstream queries keep working exactly as before.
The result: real-time data flows from your sources into governed, query-ready Iceberg tables in your own Google Cloud Storage, optimized continuously, queryable from BigQuery and beyond — and you never gave up control of a single byte.
Built on proven AWS foundations
We didn't start from zero for Google Cloud. Qlik Open Lakehouse first shipped on AWS, where it's already running these same patterns: CDC and streaming ingestion, Spot-based compute for cost-efficient optimization, Iceberg tables in customer-owned object storage, and open querying across engines.
Bringing it to Google Cloud is a deliberate step toward feature parity across clouds. The architecture mirrors what's proven on AWS, adapted to Google Cloud's building blocks: Compute Engine in place of EC2, Google Cloud Storage in place of S3, BigQuery and the Google Lakehouse runtime catalog as the native query and catalog layer. Same engine, same principles, GCP-native execution.
This matters because of what it says about our direction. We believe you should be free to choose where your data lives, which engine queries it, and to change your mind later without a migration project. Delivering the same open lakehouse capability across AWS and Google Cloud, with more to come, is how we make that freedom real rather than a slide in a pitch deck.
Why it matters
Everything above maps to three things we care about:
TRUST
Build governed data products in Qlik Talend Cloud on top of your Iceberg tables. Your data stays in open Apache Iceberg format, in your own account.
FREEDOM
Query from BigQuery, Trino, or any Iceberg-compatible engine, and swap or add engines later without re-architecting. No lock-in to a single vendor or format.
COST OPTIMIZATION
Spot VM compute and zero-copy, single-storage architecture mean you spend far less to move and hold your data at scale.
Join the Private Preview
We're accepting participants now, and we're prioritizing organizations already standardized on Google Cloud and working with BigQuery or Apache Iceberg in their accounts.
Request your spot in the Private Preview →
Opt in through the form and the Qlik team will reach out to schedule a welcome call. Already signed up? Sit tight — we'll be in touch.
...View More
The World Cup is over, Spain are champions, and the third and final episode of ourModel vs. Masterypodcast series has wrapped up with a verdict on the question we've been asking all along: when it comes to predicting football's biggest tournament, does the edge belong to the model or the human expert?
Host Adam sat down with Steve Palmer and Nick Magnuson to close out the series, and the results were more decisive than anyone on the losing end expected.
What if data quality borrowed its playbook from DevOps? Software teams learned long ago that the cheapest bug to fix is the one you never let reach production. The same principle applies to data. Learn how shifting quality and governance left, combined with agentic data engineering and observability, helps stop bad data before it impacts dashboards, AI, and business decisions.
For two decades, the job of a data engineer was deceptively simple to describe: move data from system A to system B so that someone downstream could analyze it. That era is ending. As agentic AI moves out of the lab and into core operations, a new and far more demanding consumer has arrived at the table — the autonomous agent. And agents are unforgiving about the data they are fed.
Welcome to agentic data engineering: the practice of both using and building agentic AI capabilities to deliver trusted data for AI workloads. It marks a shift from building pipelines by hand to orchestrating self-sustaining systems — and it changes what the modern data team is measured on.
Everybody wants to rule their data world — and this release gets you closer. Twelve new Stitch-powered connectors land in Qlik Talend Cloud, from Circle CI to Shopify to Microsoft Teams. But the bigger news drops on 1 July: every Stitch-originated connector now travels the full journey — Replication, Pipelines, and Qlik Open Lakehouse. No more "Replication only." Read on.
