SOFTWARE THAT MAKESYOUR HARDWAREACTUALLY WORK
IoT platforms, companion apps, and device management — built for the hardware it serves.
Ankh builds the software layer that transforms embedded hardware products into connected, manageable, monetizable systems — purpose-built for hardware-adjacent applications where IoT data pipeline constraints, device management reliability, and real-time latency requirements shape architecture. From BLE companion apps to multi-tenant SaaS platforms ingesting millions of device events per day, we build the full stack with CI/CD, containerized deployment, and SOC 2-aligned security.
What We Build
IoT Device Management Platforms
Multi-tenant SaaS platforms for fleet device management — OTA firmware deployment, device provisioning, health monitoring, alert routing, and configuration management — built on AWS IoT Core, Azure IoT Hub, or custom MQTT brokers with PostgreSQL/TimescaleDB backends, React dashboards, and RBAC.
Real-Time Sensor Data Visualization
Web dashboards consuming live device telemetry over WebSocket or MQTT with sub-second update latency — time-series charts, geographic maps, threshold alerting, and configurable data retention; TimescaleDB continuous aggregates power historical trend views over millions of sensor readings.
BLE Companion Apps (iOS & Android)
React Native and Flutter companion apps for BLE-connected hardware — device discovery, pairing, configuration interfaces, OTA firmware UX, and cloud sync — handling MTU negotiation, connection parameter tuning, and bonding flows that trip up teams without BLE-specific mobile experience.
REST & GraphQL API Development
Backend APIs in Node.js, Python (FastAPI/Django), or Go — RESTful for device-facing integrations, GraphQL for frontend querying — with JWT/OAuth 2.0 authentication and API key management for per-device revocation.
IoT Data Ingestion & Processing Pipelines
High-throughput MQTT ingestion pipelines with message routing, protocol translation (MQTT to HTTP, CoAP to MQTT), data enrichment, and fan-out to TimescaleDB, S3 data lake, and real-time alert engines — Kafka or AWS Kinesis for high-volume, NATS for low-latency small fleet architectures.
React & Vue Frontend Development
TypeScript-first React (Next.js) or Vue 3 (Nuxt) — component libraries, design system implementation, WCAG 2.1 AA compliance, and Playwright/Cypress end-to-end coverage for dashboards, configuration UIs, and customer portals.
Docker, Kubernetes & CI/CD Infrastructure
Containerized deployment on AWS ECS/EKS, GKE, or Azure AKS with Helm, horizontal pod autoscaling, and secrets via AWS Secrets Manager or Vault; GitHub Actions or GitLab CI with automated testing, image scanning, and staged deployment gates.
SOC 2 Aligned SaaS Security
SOC 2 Type II aligned architecture — AES-256 at rest, TLS 1.3 in transit, append-only audit logging, least-privilege RBAC, vulnerability scanning in CI, dependency automation, and control narratives required for the audit period.
How We Work
Technical Discovery & Architecture
Device data model, user workflows, integration requirements, and scale targets are mapped to an architecture document — component topology, data flow diagrams, API contracts, and technology selections with rationale. IoT platforms are sized for ingestion rate, storage growth, and concurrent user load.
API Contract & Data Model Design
API contracts and data models are defined in shared specifications — OpenAPI 3.0 for REST, GraphQL for query interfaces, MQTT topic hierarchy for device communication — reviewed by all stakeholders before implementation begins. Database schema covers time-series indexing, multi-tenant partitioning, and zero-downtime migration.
Sprint Development Cycles
Development runs in 2-week sprints with working software demonstrated against staging at every sprint review; the main branch stays deployable via trunk-based development with feature flags so no long-lived branches create merge complexity.
Automated Testing & Quality Gates
Every pull request is gated by a CI pipeline running Jest/Vitest unit tests, Supertest API integration tests, and Playwright/Cypress end-to-end tests with enforced coverage thresholds (80% backend, 70% frontend). Container image scanning with Trivy and static analysis block known vulnerability classes before they reach production.
Staging Validation & Load Testing
Every production deployment is validated in a staging environment that mirrors production, with k6 or Locust load tests simulating 2× peak device ingestion rate and chaos tests — container restarts, database failover, broker partition — confirming correct recovery from infrastructure failures.
Production Deployment & Observability
Deployments use canary rollout (10% traffic first) with automatic rollback on error rate or latency thresholds, backed by structured logging, OpenTelemetry tracing, and dashboards covering connection counts, message throughput, API latency, and query performance. On-call runbooks are produced for every production alert.
The Software Architecture Details That Scale
IoT Data Architecture: TimescaleDB, Partitioning, and Query Performance
Standard PostgreSQL schemas fail at IoT scale — unbounded write rates and month-spanning queries demand a different approach. TimescaleDB hypertables with chunk partitioning, continuous aggregates pre-computing 5-min/1-hr/1-day rollups, and Kafka front-loading above 10K msg/s keep p99 dashboard queries in milliseconds.
IoT ArchitectureBLE Stack Integration: The Mobile Development Details That Break Apps
iOS requires CoreBluetooth state restoration for background reconnection; Android has documented MTU bugs on Samsung devices that silently disconnect above 185 bytes. A validated BLE utility layer covers connection state machines, MTU fallback, and OTA chunking — tested against iOS 16–18 and Android 11–14 before every release.
Firmware DevelopmentSOC 2 Compliance Architecture for SaaS IoT Products
SOC 2 Type II evaluates whether controls operated effectively over a 6–12 month observation period — retrofitting before an audit is far more expensive than building in from the start. Encryption at rest, TLS 1.2+ with HSTS, append-only audit logging, RBAC, and dependency scanning (48-hour CVE SLA) are provisioned in initial Terraform so evidence accumulates from day one.
Embedded Security
IoT platforms, companion apps, and device management — built for the hardware it serves.
Software Development Across Ankh's Product Range
Every connected hardware product needs a software layer — from a simple companion app to a multi-tenant enterprise IoT platform — and our practice serves every product category we build hardware for.
Wearable Electronics
React Native companion app with BLE sync, health visualization, and HealthKit/Google Fit integration
View ProductMedical Devices
HIPAA-aligned data handling, audit logging, and clinician-facing web dashboards with patient data access controls and HL7 FHIR export
View ProductCold Chain Monitoring
Excursion alerting with multi-carrier cellular integration, PDF reports, and regulatory export formats
View ProductStructural Monitoring
Time-series waveform visualization with FFT overlays, configurable alarms, and automated structural health reports
View ProductIndustrial Equipment Monitoring
OPC-UA and Modbus TCP integration layers bridging legacy industrial equipment to modern cloud IoT platforms
View ProductAI Vision Systems
Inference streaming dashboard, annotation management for retraining curation, and model performance monitoring with drift detection
View ProductPOS & Kiosk Hardware
Remote management portal with screenshot capture, app deployment, and uptime monitoring for kiosk fleets
View ProductWireless Communication
Network topology visualization, link quality dashboards, firmware OTA management, and configuration rollout tooling for mesh network deployments
View ProductData Loggers
USB and BLE data retrieval app, cloud sync pipeline, and data export tooling for CSV, JSON, and application-specific formats
View ProductRuggedized Devices
Intermittent-connectivity-tolerant sync protocols, offline-first mobile apps, and store-and-forward message queuing for field deployments
View ProductSmart Home Automation
Apple Home / Google Home integration layers, Matter provisioning flows, and custom automation rule engine UI for prosumer products
View ProductPower Monitoring
Energy dashboard with utility rate schedule integration, cost analytics, carbon intensity overlays, and demand response event management
View ProductHardware Context in Every Software Decision
NB-IoT bandwidth limits, BLE MTU constraints, and idempotency requirements for mid-transmission power loss break architectures designed without embedded context. Our software engineers work alongside embedded engineers from the start, so the design is correct the first time a real device connects.
Production-Grade Infrastructure From the First Sprint
Unindexed databases, no observability, and SSH-only deployments surface as infrastructure debt under production load. Containerized apps, structured logging, CI/CD, and query-tuned databases are provisioned from sprint one — retrofitting under incident pressure costs far more.
End-to-End Ownership Across Hardware and Software
MQTT payload schemas, the firmware client that serializes them, and the backend that deserializes and stores them are owned by the same team — so malformed packets are fixed at the root rather than papered over across three codebases.
Deliverables
Software Architecture Document
Component topology diagram, data flow documentation, technology selection rationale, API contract specifications, and database schema design with indexing strategy.
Frontend Application
Production-deployed React or Vue application with TypeScript, design system implementation, WCAG 2.1 AA compliance, and Playwright/Cypress end-to-end test suite.
Backend API Service
Containerized Node.js, Python, or Go API service with OpenAPI documentation, JWT/OAuth 2.0 authentication, rate limiting, and full integration test coverage.
Mobile Companion App
React Native or Flutter iOS and Android app, submission-ready for App Store and Google Play, with BLE stack integration, offline data handling, and OTA update support.
IoT Data Pipeline
MQTT ingestion service, message routing configuration, TimescaleDB hypertable schema with continuous aggregates, and data retention policy configuration.
CI/CD Pipeline & Infrastructure
GitHub Actions or GitLab CI pipeline with automated test gates, container image scanning, Terraform infrastructure-as-code, and staged deployment configuration.
Automated Test Suite
Unit, integration, and end-to-end test suites with coverage reporting, plus k6/Locust load test scripts validated against production-scale device fleet simulation.
Operations Runbook & Observability Setup
Grafana/CloudWatch dashboards for application health metrics, structured logging configuration, alert thresholds, and on-call runbooks for every production alert.
Deliverables scale with engagement scope — a companion app project has different outputs than a multi-tenant IoT platform program.
Often Paired With Software Development
Firmware Development
Co-owning firmware and backend means MQTT schemas, serialization, and ingestion are designed as a unified system rather than thrown over a wall.
ExploreIoT Architecture
A dedicated architecture engagement defines connectivity topology, protocol selections, and cloud platform strategy before development begins.
ExploreAI/ML Development
ML inference serving requires streaming pipelines, model versioning, and feedback loops built into the application architecture from the start.
ExploreEdge Data Pipelines
Edge processing reduces ingestion costs for applications that filter or aggregate before transmission — requiring firmware and cloud software co-design.
ExploreWork That Demonstrates the Standard
Multi-Tenant IoT Platform for Industrial Sensor Network — 8,000 Devices
A Go routing service and TimescaleDB with continuous aggregates sustained 12,000 msg/s across 8,000 sensor nodes, cutting p99 dashboard queries from 4.2 s to 85 ms. Staged OTA in 500-device batches reduced the fleet update cycle from 3 days to 4 hours.
Read the Case StudyBLE Companion App for Clinical Wound Assessment Device
A custom BLE state machine handled iOS background reconnection and MTU negotiation for 1–2 MB image transfers, with HIPAA-compliant cloud sync and HL7 FHIR R4 EHR integration. The app launched at 4.8 stars with zero BLE connection failures in 90 days of clinical use.
Read the Case StudyReady to Build the Software Your Hardware Deserves?IoT platforms, companion apps, and device management — built by engineers who understand hardware.
Ankh builds software that works with your devices from the first connection — from architecture to production deployment to SOC 2 compliance, owning the full software stack.