Samleng — AI voice notifications for enterprise logistics

How Ongkrong Consulting designed and built an automated native-language calling platform that cut a major logistics network's customer-notification cost by roughly 75% — on the order of US$3M a year.

Public, de-identified version. Client de-identified; no pricing, margins, or stack internals. Figures are engagement design figures, validated pre-rollout. The full, named technical version is available to qualified prospects under NDA.

At a glance

Client

One of Southeast Asia's largest logistics and parcel networks (named under NDA)

Sector

Logistics / last-mile parcel delivery

Challenge

A manual, branch-based operation calling customers by hand to announce parcel arrivals

Scope

Discovery → solution design → telecom procurement → financial modelling → cloud architecture → build & load test → production rollout plan

Delivery model

AI-native: a lean senior team directing AI agent fleets under human architecture and review

Headline outcome

~75% reduction in notification cost — on the order of US$3M per year

Engineering proof

Platform passing 181/181 automated tests with zero dependency vulnerabilities; outbound dialling layer load-tested to 20,000 concurrent calls

Timeline

~7 weeks from first engagement to a built, validated platform and a contract-ready rollout

Summary

A market-leading logistics network was notifying customers of parcel arrivals entirely by hand — hundreds of branches, a calling workforce in the thousands, hundreds of thousands of calls a day. The cost ran into the millions annually, and an internal attempt to automate it had stalled.

Ongkrong Consulting designed and built Samleng, a platform that places parcel-arrival calls in natural, native-language voice, triggered automatically the moment a parcel is scanned at a branch. The design cuts the cost of the operation by roughly 75% — around US$3M a year. The engagement was full-scope, not just software: opportunity discovery and business-case construction, a solution scoped against a complete call-flow audit, live telecom procurement to secure a local caller ID, a defensible cost-and-margin model, a documented enterprise cloud architecture, and a production-grade governance package.

The challenge

For a large parcel network, the “your parcel has arrived” call is part of the core service. Done by hand across hundreds of branches, it ties up a calling workforce in the thousands and costs millions a year — while still being slow, inconsistent, and impossible to measure. The cost was large, recurring, and entirely manual; an internal effort to automate it had not delivered. The question was not whether to automate, but who could actually ship it.

The approach

1. 01

   Engagement and discovery

   We rebuilt the cost baseline from the ground up — branch count, staffing, per-agent cost, and a defensible regional loading multiplier — so the business case would survive a CFO's scrutiny rather than rely on a headline figure. We mapped how the decision would actually be made and where the real competition was: the client's own internal build, which had already lost months.

2. 02

   Solution scoping and MVP

   We scoped an event-driven product: a parcel is scanned, an event reaches the platform, a native-language voice message is rendered and the call is placed within seconds, and the outcome is tracked. Scoping included a full end-to-end call-flow audit rather than a happy-path demo — a multi-agent review that surfaced 8 critical, 13 high, and 12 medium failure modes, each given a chosen fix before any real customer was dialled. A separate adversarial red-team pass found and triaged 61 issues into a prioritised hardening backlog.

3. 03

   Telecom procurement

   Placing automated calls with a genuine local caller ID is a telecom problem, not a coding one — a non-local number collapses pickup rates. We ran a live procurement across the call-platform layer and the local carrier market, took a competitive carrier offer to a signed quotation, and designed a multi-vendor path so no single provider owns the whole stack. The same process surfaced the single commercial variable that governs the platform's economics — and flagged it for written resolution before any full-volume commitment.

4. 04

   Cost and financial model

   We built a model designed to survive a CFO and to tell us the truth about our own economics, running multiple cost scenarios rather than one flattering case. It proved the client's ~75% saving and isolated the one variable that could undermine it — early, while it was still cheap to address. The model made plain that infrastructure cost is a rounding error next to telephony.

5. 05

   Cloud architecture

   We produced an enterprise target architecture documented to a standard a buyer's technical reviewer can audit: ten formal Architecture Decision Records, each stating the decision, the alternatives, and the tradeoff; service-level objectives engineered to 99.9% while publishing a 99.5% SLA, with a 1-hour recovery-time objective and a 5-minute recovery-point objective. The design and its cost were validated with the platform vendors' own solution engineers, then pressure-tested through structured review and an independent automated audit.

6. 06

   Deployment and load test

   The platform passes 181 of 181 automated tests with zero dependency vulnerabilities. The outbound dialling layer was soak-tested to 20,000 concurrent calls — roughly 250× the launch ceiling — holding tens of thousands of call-creations per second at a 99th-percentile latency of 76 milliseconds, with no errors. A reusable load suite was built for every future scale step.

7. 07

   Production rollout and governance

   The enterprise wrapper is complete and contract-ready: a phased, gated migration plan with explicit go/no-go criteria and a parallel-run period before cutover; a 25-item risk register, severity-rated with named owners and mitigations; and a governance layer including a client-facing data-security overview and a pilot security checklist gating real customer traffic.

Results

• ▸~75% reduction in the cost of customer notification — on the order of US$3M a year.
• ▸A built, tested platform: 181/181 automated tests passing, zero dependency vulnerabilities, and an outbound layer proven to 20,000 concurrent calls.
• ▸A working local-caller-ID telephony path secured through real carrier procurement and a signed quotation — solving a problem off-the-shelf providers could not.
• ▸A decision-grade financial model that proved the saving and isolated the one variable governing profitability.
• ▸An audit-ready architecture package — ten ADRs, SLOs, and a 25-item risk register — validated with the platform vendors' solution engineers.
• ▸Out-executed an internal build that had stalled, reaching a built-and-validated platform with a staged rollout in about seven weeks.

Why it worked

Capabilities demonstrated

• ▸Opportunity discovery and business-case construction that survives CFO scrutiny
• ▸Telecom and vendor procurement, including carrier and SIP negotiation in emerging markets
• ▸Financial and margin modelling that isolates the variables that matter
• ▸Audit-ready cloud solution architecture (decision records, SLOs, risk registers)
• ▸Production engineering and load testing of real-time, high-concurrency voice systems
• ▸Enterprise security and governance packaging

Prepared by Ongkrong Consulting. Client details available to qualified prospects under NDA. Accurate as of June 2026.