RESPONSIBILITY

AI Ethics

We believe AI tools that influence careers must be built with transparency, fairness, and human agency at the centre.

Transparency in AI Output

atsprecise uses large language models (LLMs) to analyse resumes and job descriptions. We surface explicit reasoning for every score and recommendation — not just numbers. You can read why a skill was flagged, what keyword was missed, and how the model assessed your experience against the role.

We never present AI scores as objective truths. Every evaluation includes a confidence context and should be read as an advisory signal, not a definitive judgement.

Fairness & Bias Awareness

Language models can reflect biases present in their training data, including demographic, cultural, and linguistic biases. We acknowledge this limitation openly:

We do not use or expose information about a candidate's gender, ethnicity, age, or other protected characteristics.
We do not score candidates against each other — evaluations are always relative to a specific job description.
We encourage users to treat AI recommendations as one input among many, not a final hiring or rejection criterion.

If you encounter a response that appears biased or harmful, please report it to lbharath0712@gmail.com.

Human Agency

atsprecise is a tool to augment your judgement, not replace it. We design every feature so that you remain in control:

AI-generated resume tailoring suggestions are always editable — we never auto-apply changes.
Your profile data is yours; you can export or delete it at any time.
We do not make automated hiring decisions. No employer data is shared through the platform.

Data Minimisation

We process only the data necessary to provide the requested analysis. Resume and job description text is transmitted to our AI provider (OpenRouter / Google Gemini) for inference and is not retained for model training by our instruction. We do not build behavioural profiles or add tracking beyond what is required for session management and rate limiting.

Accountability

We are committed to continuous improvement of our AI systems. If our models produce harmful, misleading, or inaccurate outputs at scale, we will investigate and take corrective action. We welcome scrutiny and feedback from users, researchers, and the broader community.

← Back to home