IT Manager and hands-on engineer, Sri Lanka

I keep critical systems running.
Then I build what should exist next.

My work moves between networks, servers, backup and recovery, Android internals, and the web. The common thread is simple: understand the system, remove the friction, and make the result dependable.

Nadeeja Nirmala
Current focusReliable operations
Polonnaruwa, Sri Lanka

A wider engineering view

The strongest technology work is not a screenshot. It is a system people can trust.

I support the infrastructure behind day-to-day operations and build products where the interface needs to explain what the technology is doing. That combination has taught me to care about the quiet details: failure paths, recovery time, access boundaries, accessibility, and whether the person using the result can make sense of it.

24/7 operations3-2-1 recovery designAndroid 11 to 17 platform workWeb from code to motion

Selected work

One portfolio.
Five sides of the craft.

Each chapter is a different kind of engineering problem. Together they show how I think, build, protect, and communicate.

01 Android product design

NIRALYTH

A local-first music player concept made to feel as precise as the audio it plays.

Lossless playback, dual analog VU meters, lyrics, artwork, and a precision equalizer come together in one focused Android experience. The public site is a product showcase; the application source and backend remain private.

View the live showcase
barrylk.github.io/niralyth-showcase
NIRALYTH music player product showcase
InterfaceAudio with a pulse

02 Android telephony engineering

Pixel IMS 5G

A clearer path into the radio settings Android keeps out of sight.

Built for Tensor-powered Pixel devices, the utility opens IMS and radio configuration paths through root or Shizuku, reports live VoLTE and VoWiFi status, and supports signed releases with OTA-safe update metadata.

Explore Pixel IMS 5G
Pixel IMS 5G Android application screens
Radio stateIMS registered
5Glive diagnostics

03 Web experience engineering

The web is part of the product.

Technical depth should feel clear, responsive, and worth exploring.

I build semantic, responsive sites that turn dense engineering work into a focused story. The system behind this portfolio uses static HTML for resilient discovery, purposeful GSAP motion, accessible interaction, and no front-end framework.

Read the web engineering case study
nirmala.is-a.dev

BUILDING ACROSS LAYERS

Clear systems.
Human interfaces.
Explore the work

04 Disaster recovery

Recovery should be ready before it is needed.

A 3-2-1 architecture designed around the systems that production depends on.

ERP and production data are protected across local RAID, cloud storage, and encrypted off-site media. Recovery targets are defined, integrity is checked, and the path back to operation is treated as part of the design.

Read the recovery case study

05 Network engineering

The right traffic, on the right path.

A manufacturing network redesigned around separation, visibility, and production priority.

Production, office, and CCTV traffic sit in distinct security zones. Hardened boundaries, encrypted DNS, intrusion detection, and ERP-priority QoS reduce unnecessary exposure without getting in the way of work.

Read the network case study

06 Systems engineering

Hybrid systems, operated as one.

Windows, Linux, and cloud services connected without losing sight of security or maintainability.

The work spans bare-metal deployment, hardening, service isolation, remote administration, identity, and cloud integration. Monitoring turns the environment into something observable before a user has to report a problem.

Read the systems case study

07 Android privacy engineering

Lumen

Private connectivity that feels calm, direct, and understandable.

Lumen provisions full-device IKEv2/IPsec tunnels through Android's platform VPN stack. Secure profile import and system-managed credentials keep the workflow focused while supporting Android 11 through 17.

View the Lumen showcase
Lumen privacy-focused Android VPN application
Tunnel stateProtected
IKEv2platform VPN

How I work

Evidence first.
Always verify the result.

The same method works for a broken route, a failed restore, or an Android service that behaves differently on a new release.

  1. 01
    Observe

    Read the real state before touching it.

  2. 02
    Isolate

    Separate symptoms, dependencies, and risk.

  3. 03
    Engineer

    Build the smallest durable path forward.

  4. 04
    Verify

    Test the happy path, the failure path, and recovery.

Open to the right opportunity

Looking for someone who can see the whole system?

I would like to hear about roles where infrastructure judgment, hands-on engineering, and thoughtful product work belong in the same conversation.

Start the conversation