AI / ML Engineer

Kaaliber; Dadri; On-site; INR 4,200,000 - 6,000,000; Full time

Job summary

Kaaliber is seeking an experienced AI/ML Architect to define architecture across products, lead LLM workflow design, and ensure secure, scalable model deployments. You will mentor engineers and coordinate with backend, frontend, QA, security and cloud teams to deliver reliable AI solutions.

The role emphasizes responsible model evaluation, risk mitigation, and clear communication of trade-offs to leadership. Strong Python, ML foundations, and production ML experience are essential.

Qualifications

• Advanced Python and strong foundations in machine learning, statistics and experimental design.
• Proven ownership of production LLM applications, including structured outputs, tool calling, retrieval and evaluation.
• Experience delivering classical ML systems using scikit-learn, PyTorch, TensorFlow or comparable frameworks.
• Strong understanding of precision, recall, ranking metrics, calibration, drift, bias, data leakage and error analysis.
• Experience with model deployment, versioning, feature or data pipelines, observability, and rollback.
• Ability to architect secure AI services and defend against prompt injection, unsafe tool use and sensitive-data leakage.
• Strong SQL and data-analysis skills with experience working on event, behavioural, or time-series datasets.
• Technical leadership, mentoring, design review and stakeholder communication experience.

Responsibilities

• Define AI/ML architecture, model-selection strategy, reusable service patterns and technical standards across products.
• Lead the design of LLM workflows, retrieval systems, tool use, routing, structured outputs and human-review controls.
• Guide predictive-model development for classification, ranking, recommendation, forecasting and anomaly detection.
• Establish dataset, experimentation, offline evaluation, online measurement and model-releasepractices.
• Define quality, safety, latency and cost thresholds and ensure model changes are validated before production release.
• Design MLOps and LLMOps capabilities covering versioning, deployment, tracing, drift, rollback and provider resilience.
• Review AI designs and code, create reference implementations and resolve complex modelling or production issues.
• Mentor AI/ML engineers, improve technical depth and coordinate delivery with backend, frontend, QA, security and cloud teams.
• Assess external models and platforms while protecting data, limiting vendor dependence and controlling operational cost.
• Communicate evidence, uncertainty, model limitations and technical trade-offs clearly to leadership and delivery teams.

Skills

Advanced Python

ML Foundations

Experimental design

Production LLM apps

ML frameworks

SQL & data analysis

Security & data privacy

Mentoring & leadership

Stakeholder communication

Tools

scikit-learn

PyTorch

TensorFlow

Kubernetes

CUDA

Job description

Responsibilities:

• Define AI/ML architecture, model-selection strategy, reusable service patterns and technical standards across products.
• Lead the design of LLM workflows, retrieval systems, tool use, routing, structured outputs and human-review controls.
• Guide predictive-model development for classification, ranking, recommendation, forecasting and anomaly detection.
• Establish dataset, experimentation, offline evaluation, online measurement and model-releasepractices.
• Define quality, safety, latency and cost thresholds and ensure model changes are validated before production release.
• Design MLOps and LLMOps capabilities covering versioning, deployment, tracing, drift, rollback and provider resilience.
• Review AI designs and code, create reference implementations and resolve complex modelling or production issues.
• Mentor AI/ML engineers, improve technical depth and coordinate delivery with backend, frontend, QA, security and cloud teams.
• Assess external models and platforms while protecting data, limiting vendor dependence and controlling operational cost.
• Communicate evidence, uncertainty, model limitations and technical trade-offs clearly to leadership and delivery teams.

Requirements:

• Advanced Python and strong foundations in machine learning, deep learning, statistics and experimental design.
• Proven ownership of production LLM applications, including structured outputs, tool calling, retrieval and evaluation.
• Experience delivering classical ML systems using scikit-learn, PyTorch, TensorFlow or comparable frameworks.
• Strong understanding of precision, recall, ranking metrics, calibration, drift, bias, data leakage and error analysis.
• Experience with model deployment, versioning, feature or data pipelines, observability, and rollback.
• Ability to architect secure AI services and defend against prompt injection, unsafe tool use and sensitive-data leakage.
• Strong SQL and data-analysis skills with experience working on event, behavioural, or time-series datasets.
• Technical leadership, mentoring, design review and stakeholder communication experience.

Preferred experience:

• Multi-model routing, agent orchestration, vector retrieval and knowledge-graph experience.
• Causal inference, uplift modelling, recommendations, optimisation or contextual bandits.
• Multimodal models involving text, image, audio or video.
• Cloud-based ML platforms, Kubernetes, distributed inference and GPU workload optimisation.
• Experience building shared AI platforms or standards used by more than one product team.