Phishing Awareness

Extending Gophish for AI-Phishing Content Generation

Monthly phishing simulations need fresh, convincing content. Here's how Halodoc built an AI generator into Gophish to create varied pretexts in English and Indonesian.

AI Agent

One Platform for Every AI Agent at Halodoc

Several Halodoc teams built AI agents on their own, each with its own guardrails and infra. Here's how we brought them onto one shared, governed agent platform.

Web Development

No Green, No MR: How We Ship Core Web Vitals Fixes with an Agentic Workflow

How Halodoc built an agentic, ship-gated workflow to fix Core Web Vitals on Angular apps — with a human confirming before every MR opens.

JDK25

Scaling JDK 25 Migration with AI at Halodoc

Halodoc's AI skill migrated 140 Java services to JDK 25 end to end, resolving every exception without a hand-edited file.

CI-CD

Automating Root Cause Analysis for CI/CD Failures Using AI at Halodoc

At Halodoc, our CI/CD pipelines run ~800 builds a day, with ~190 failures needing manual log digging. This post covers how we built a three-tier Root Cause Analysis agent — rules, a signature cache, and AI — that instantly triages failures and posts fixes to Google Chat.

Data Engineering

DE Central: Halodoc’s AI Agent for Data, Pipelines, and Code

DE Central is Halodoc's new AI agent that makes the Data Platform easy for everyone — diagnosing pipelines, answering data questions, and tracing table usage in one place.

automation

How We Automated Android Memory Leak Triage — From LeakCanary to Merge Request with AI

How Halodoc turned LeakCanary's on-device alerts into an AI pipeline that files JIRA tickets and opens reviewable merge requests for Android memory leaks.

Angular Update

How We Automated Angular Upgrades at Halodoc using AI

How Halodoc automated Angular upgrades across ~40 apps with a single Claude AI skill — grounded in Angular's own migration data, with a human review gate.

iOS

How We Automated iOS Crash Triage with AI — Without Letting It Merge

How we turned crash triage from an overnight fire drill into an automated pipeline that hands engineers a review-ready fix — without ever letting AI merge on its own.

Cyber Security

From SIEM Alerts to Action: Building an AI-Powered SOC with Amazon Bedrock Agents

Manual Tier-1 alert triage doesn't scale. We built an AI-driven triage layer on Amazon Bedrock Agents that investigates every SIEM alert in seconds and returns a structured verdict. Plus the counterintuitive result from our model evaluation.

Apache Airflow

Leveraging Data Assets features in Airflow 3.0 to optimise resource utilization by more than 30%

How Halodoc cut Airflow worker CPU by over 70% using Data Assets and the deferrable Redshift operator, plus the bugs we hit upgrading to Airflow 3.2.1 along the way.

Jenkins

Beyond Executor Slots: Building a Memory-Budget Allocator for Jenkins CI

How Halodoc built a memory-budget allocator for Jenkins to stop OOM kills on shared Mac CI workers and make builds reliable again.

AI

Automating Code Reviews with AI: How Halodoc Scaled Quality Across the Full Stack

How Halodoc built an AI-powered code reviewer into its Jenkins pipeline to catch issues instantly and scale review quality.

AI Skills

How We Turned Data Engineering Runbooks Into Reliable AI Skills

Building AI skills that hold up in production isn't about the model — it's about the structure you put around it. This blog breaks down how we put in the implementation layer behind grounding, structured generation, and human-in-the-loop checkpoints.

SEO

Automating Dynamic Prerendering at Halodoc: A Data-Driven Approach to SEO and Faster Page Discovery

We stopped guessing which pages to prerender. By connecting the Google Search Console API directly to our build pipeline, Halodoc now automatically optimizes SEO based on real-time search trends. Here's how we did it. 🚀

Cloud Security

From Single Point of Failure to Warm Standby: A Practical Core Switch Disaster Recovery Playbook

Overview Every organization depends on its network to keep operations running. But for us, the stakes were higher than most. Our core business operations, internal systems, communications and cross-team workflows run entirely on this network. On top of that, our Customer Service team operates 24/7. Every minute of downtime

LLM

Reducing LLM Token Costs by 15% by Switching from JSON to TOON Format

Introduction At Halodoc, LLMs power multiple high-volume healthcare workflows - from clinical decision support and document processing to intelligent routing and classification. Scaling these AI capabilities responsibly means keeping a close eye on efficiency, and one of the biggest levers we found was in how we format the data we

API Security Testing

Securing the SDLC: A Zero-Trust Framework for Supply Chain Security

Managing CVEs across polyglot microservices (Java, Python, Node, Go) was chaotic. Halodoc built a 3-layer supply chain security model — creating one source of truth with security gates from PR to image registry.

App Development

How We Generate Production SwiftUI from Figma Using AI Skill at Halodoc

At Halodoc, we've made AI central to mobile development by building FigmaToSwiftUI—a Skill that generates production-ready SwiftUI code directly from Figma designs. Powered by Figma MCP and Code Connect, this approach cuts repetitive work and ships features faster. Here's how we did it.

Data Engineering

Enabling Data Intelligence: Data Profiling Framework at Halodoc

At Halodoc, we built a scalable Data Profiling Framework to bring visibility and trust to data pipelines. Discover how our Airflow-native system handles column profiling, join intelligence, and source analysis across hundreds of tables—making data quality a core engineering concern.

Angular

Migrating the Halodoc Website to Zoneless Angular

Halodoc migrated Angular's SSR frontend from Zone.js to zoneless change detection, eliminating unpredictable server errors and SSR failures. The result: a cleaner bundle, more debuggable code, and better alignment with modern Signals reactivity—all measured in production.

Data Engineering

Building Self-Healing Data Pipelines at Halodoc

At scale, data powers everything — from executive dashboards and operational analytics to financial reconciliation and customer-facing features. But as data platforms grow, one reality becomes unavoidable: pipelines break. What if the platform could heal itself?

Cyber Security

Building Secure PII Encryption at Scale

Halodoc addresses the challenge of securing PII at scale using a config-driven, application-level encryption architecture. By using AES-256-GCM and Vault-cached keys, they achieved transparent data protection across Java and Go services with negligible latency and zero-downtime migration.

KRaft

Migrating AWS MSK from ZooKeeper to KRaft: A Canary Approach

This blog post shares our end-to-end journey of migrating Amazon MSK from ZooKeeper to KRaft, covering validation, migration strategy, execution, rollback planning, and key learnings.

Android Development

Migrating Apps to Android 16: Our Journey at Halodoc

This blog shares how we addressed the Android 16 platform changes, the challenges we encountered, and how we ensured a smooth transition without compromising the user experience.