RESEARCH

ADVANCING COGNITIVE ARCHITECTURES

Collaborating with leading researchers to push the boundaries of intelligent testing

RESEARCH FRONTIERS

[AI-01]

AI-POWERED TESTING

Exploring machine learning models for intelligent test generation, self-healing tests, and predictive bug detection.

[TS-02]

TEST AUTOMATION

Developing novel approaches to reduce test maintenance overhead and increase coverage in modern web applications.

[QA-03]

QUALITY ASSURANCE

Researching metrics, methodologies, and best practices for measuring and improving software quality at scale.

PUBLISHED RESEARCH

// LITERATURE SURVEY · 2026

Energy Efficient and Carbon Aware Resource Scheduling for Large Language Model Workloads in Cloud Computing: A Comprehensive Survey

A comprehensive investigation into green computing paradigms for artificial intelligence. This survey maps carbon-intelligent scheduling algorithms, dynamic power management, and hardware-software co-optimization strategies specifically tailored for orchestrating LLM training and inference workloads across distributed cloud infrastructure.

Fawad Ul Haq · literature survey · not peer-reviewed · adjacent to, not about, SOVEREIGN

↓ DOWNLOAD PDF (223 KB)

OPEN PROBLEMS

What we are actively working on. No results to report yet — when there are, they will appear here with methodology and numbers you can check.

01 // Locator healing via semantic similarity: measuring how often a repaired selector matches author intent versus silently passing the wrong element.
02 // Whether small local models (7–14B, via Ollama) can match hosted frontier models on test-step generation, so BYOK becomes optional rather than required.
03 // Quantifying the false-positive rate of automated SQLi/XSS fuzzing against real applications, rather than deliberately vulnerable test targets.

[STATUS: IN PROGRESS · NO PUBLISHED RESULTS YET]

OPEN TO COLLABORATION

We do not currently have formal partnerships to announce. These are the three areas where we would most value collaborators — if that is you, get in touch.

[AC-4]

UNIVERSITIES

Seeking academic collaborators for empirical studies on locator healing and AI-assisted test generation. Datasets and tooling provided.

[IN-5]

INDUSTRY

Looking for QA teams willing to run SOVEREIGN against a real suite and let us publish the measured before/after, including where it performed worse.

[OS-6]

OPEN SOURCE

SOVEREIGN builds on Playwright and Ollama. We intend to open-source the locator healing layer once its behaviour is documented well enough to be useful.

COLLABORATE WITH US

Interested in collaborating on research or sharing your own work? We'd love to hear from you.

GET IN TOUCH