Teh Tarik Digital was founded in 2016 by engineers who came from international cryptocurrency exchange operations, BNM-adjacent banking infrastructure, and government technology initiatives at the ministerial level. Regulated industries, cross-border operations, high-accountability environments — that's the background behind every engagement we take on, and it's the same lens we bring to Indonesia.
Jakarta is where that matters. The city drives an estimated 70–80% of Indonesia's AI workloads and data-centre capacity, and 2026 has brought a dense run of AI-focused events to the city. The market is moving fast — so is the regulatory environment. Our own presence is in South Jakarta, where we run production work for an enterprise client through a local branch at Infinia Park, close to the districts driving much of this activity.
Indonesia's Personal Data Protection Law (UU PDP, Law No. 27 of 2022) has been fully enforceable since October 2024 — administrative sanctions of up to 2% of annual revenue apply, and 72-hour breach notification is required. But the dedicated PDP supervisory agency the law calls for is still being established, and several implementing regulations remain in draft. We think that's worth saying plainly rather than glossing over: any AI system handling personal data in Indonesia today needs to be built to the law's existing requirements, with a clear eye on what's still being finalised.
Two products, adapted rather than invented from scratch — and we're direct about where that adaptation is complete versus still a scoping conversation. That distinction matters more here than it would on a Malaysian page: an Indonesian buyer evaluating a Malaysian vendor should be able to tell exactly which claims are proven and which are being scoped, without having to dig for the caveat in the fine print.
AI workspace and RAG-based knowledge assistant, applying the same architecture proven with Malaysian enterprise clients. Current data-residency runs on Malaysian cloud infrastructure — in-country Indonesian residency is a scoping conversation, not an off-the-shelf claim.
Approval-chain and procurement orchestration, relevant to Indonesia's own enterprise and public-sector workflow challenges, with deployment architecture scoped per engagement.
Why Teh Tarik Digital for an Indonesia engagement: we're not claiming a broader Indonesian deployment track record than we have. What we bring is a founding team with real experience operating in regulated, cross-border, high-accountability environments, products already proven inside one of the region's stricter compliance regimes (Malaysia's PDPA), and an active South Jakarta client relationship. Entering Indonesia means adapting that architecture to UU PDP specifically, not assuming one compliance framework maps onto another.
Nothing here is a first build. It's proven architecture being adapted to a new regulatory environment, honestly scoped — which is a different claim than "we've deployed in Indonesia before," and we're not going to blur that line to make the pitch sound stronger than it is.
Compliance posture gets scoped before architecture — not after. That ordering matters specifically in Indonesia right now, where UU PDP is active law but its implementing regulations are still being finalised; building to a moving compliance target means starting from where the law actually stands today, not where a generic playbook assumes it does.
Understand your workflow and current UU PDP compliance posture — this determines everything downstream, so we don't skip past it to get to a faster-sounding pitch.
Architecture proposal, with data-residency conversation scoped per engagement — where the data actually lives gets decided here, in writing, not assumed and discovered later.
JomWork/IRIS adapted to your systems — adapted, meaning the underlying engine already runs elsewhere; what changes here is the regulatory tuning and the integration surface.
Live deployment with ongoing support — from a team that's reachable across the border, not a satellite office that goes quiet after go-live.
One of our own builds, documented in full — the architecture, the numbers, and the reasoning behind them.
RAG + LangGraph + HITL escalation for a high-volume e-commerce support queue, with the reasoning behind every architecture decision. Full case study →
The questions that come up first when a Malaysian vendor pitches an Indonesian deployment — answered without overstating what's proven yet. If a question isn't on this list, it's usually because the honest answer is "ask us directly" rather than something we could turn into a confident-sounding line of copy.