AI Researcher

Virallens; Bengaluru; On-site; INR 1,800,000 - 3,200,000; Full time

Job summary

Virallens is seeking an AI Researcher to design, experiment with, and advance AI/ML systems powering our products and enterprise apps. The role spans Generative AI, LLMs, CV, multimodal AI, RL, and more, with emphasis on deep expertise in at least one area and applying fundamentals to new problems.

You will reproduce research, build prototypes, evaluate rigorously, and translate promising ideas into practical capabilities, collaborating with cross-functional teams to deliver impact.

Qualifications

• Minimum 3 peer-reviewed research publications in Q1 journals and/or A conferences, through any qualifying combination.
• The publications must represent substantive research contributions in Artificial Intelligence, Machine Learning, Deep Learning, Computer Science, or a closely related field.
• Candidates will be evaluated primarily on the quality, relevance, originality, and depth of their research contributions.

Responsibilities

• Identify and investigate meaningful research problems in your area of AI/ML specialization.
• Research emerging techniques across AI, machine learning, deep learning, and related fields.
• Study, reproduce, and extend ideas from recent research papers.
• Formulate research questions, hypotheses, experiments, and evaluation methodologies.
• Design and implement novel algorithms, architectures, and AI techniques.
• Build experimental prototypes to validate research ideas.
• Conduct systematic experiments, ablation studies, and comparative evaluations.
• Analyze experimental results and identify opportunities for improvement.
• Investigate methods for improving model accuracy, efficiency, robustness, generalization, and reliability.
• Document research findings and communicate technical insights to engineering and product teams.

Job description

Employment Type: Full-time

About the Role

As an AI Researcher, you will research, design, experiment with, and develop advanced Artificial Intelligence and Machine Learning systems that power our AI products and enterprise applications.

This is a broad AI research role, not limited to NLP or LLMs. We are looking for researchers with strong expertise in one or more AI/ML research domains, such as Generative AI, LLMs, Computer Vision, Multimodal AI, Reinforcement Learning, Recommendation Systems, Search, Speech and Audio AI, Time-Series Modeling, Graph ML, Robotics, or other emerging areas.

You are not expected to have expertise across all these domains. Instead, you should have deep research expertise in at least one area and be capable of applying strong AI/ML fundamentals to new and interdisciplinary problems.

You will investigate new techniques, reproduce and extend research, formulate hypotheses, build experimental prototypes, conduct rigorous evaluations, and translate promising research into practical AI capabilities.

The ideal candidate is a research-oriented AI/ML professional who can go beyond integrating existing APIs and demonstrate the ability to identify research problems, implement ideas from papers, design rigorous experiments, publish high-quality research, and translate research into real-world impact.

What You'll Do

AI Research & Experimentation

• Identify and investigate meaningful research problems in your area of AI/ML specialization.
• Research emerging techniques across AI, machine learning, deep learning, and related fields.
• Study, reproduce, and extend ideas from recent research papers.
• Formulate research questions, hypotheses, experiments, and evaluation methodologies.
• Design and implement novel algorithms, architectures, and AI techniques.
• Build experimental prototypes to validate research ideas.
• Conduct systematic experiments, ablation studies, and comparative evaluations.
• Analyze experimental results and identify opportunities for improvement.
• Investigate methods for improving model accuracy, efficiency, robustness, generalization, and reliability.
• Document research findings and communicate technical insights to engineering and product teams.

Research Specialization

Candidates should have strong research expertise in at least one AI/ML domain, including but not limited to:

• Large Language Models and Generative AI
• Computer Vision
• Natural Language Processing
• Recommendation Systems
• Search and Information Retrieval
• AI Agents and Autonomous Systems
• Speech and Audio AI
• Time-Series Modeling and Forecasting
• Robotics and Embodied AI
• Optimization and Learning Algorithms
• Other emerging AI/ML research areas

Deep expertise in one specialization is preferred over superficial knowledge across multiple areas.

No candidate is expected to be an expert in all of the above domains.

Foundation Models and Generative AI

Where relevant to your research specialization:

• Research and experiment with foundation models and modern neural architectures.
• Explore LLMs, transformer architectures, multimodal models, and generative models.
• Investigate prompting, fine-tuning, instruction tuning, model adaptation, distillation, and efficient inference.
• Research Retrieval-Augmented Generation, semantic retrieval, and knowledge systems.
• Explore AI agents, tool use, planning, memory, reasoning, and multi-agent architectures.
• Evaluate open-source and proprietary models for specific applications.
• Develop techniques to improve factuality, reasoning, robustness, and reliability.

Candidates specializing in other AI/ML domains are not expected to have expertise in all Generative AI techniques listed above.

Computer Vision and Multimodal AI

For candidates specializing in vision or multimodal AI:

• Research computer vision and visual representation learning techniques.
• Explore image understanding, detection, segmentation, classification, generation, and visual reasoning.
• Research vision-language and multimodal foundation models.
• Develop multimodal retrieval, reasoning, and generation systems.
• Explore image, video, and cross-modal understanding.

Model Training and Evaluation

• Build datasets and experimental pipelines for model training, fine-tuning, and evaluation.
• Design appropriate benchmarks, metrics, and evaluation methodologies for your research domain.
• Perform quantitative and qualitative analysis of model performance.
• Conduct error analysis and investigate model failure modes.
• Perform ablation studies and controlled experiments.
• Compare approaches across relevant quality, computational, latency, scalability, and efficiency metrics.
• Develop reproducible research and evaluation pipelines.

Applied AI Research

• Translate research concepts into practical AI capabilities and prototypes.
• Research intelligent systems for problems such as search, recommendation, prediction, classification, reasoning, optimization, perception, and decision-making.
• Develop domain-specific AI systems.
• Identify promising academic research and evaluate its applicability to real-world problems.
• Collaborate with AI Engineering, Software Engineering, Product, and Platform teams to transition successful research prototypes into production.

Robust and Reliable AI

Depending on the candidate's research domain:

• Investigate AI robustness, reliability, interpretability, and explainability.
• Evaluate models for failure modes, hallucinations, bias, adversarial behavior, and other risks.
• Develop appropriate evaluation methodologies and guardrails.
• Research methods for improving model transparency and reliability.
• Contribute to responsible AI research practices.

Research and Publications

• Identify novel and impactful research problems.
• Develop research hypotheses and novel approaches.
• Publish research in high-quality journals and conferences.
• Present research findings to internal and external technical audiences.
• Track developments from leading AI research labs, conferences, journals, and open-source communities.
• Contribute to research publications, patents, technical reports, and open-source projects where applicable.

Required Qualification

Minimum 3 peer-reviewed research publications in Q1 journals and/or A conferences, through any qualifying combination.*

The publications must represent substantive research contributions in Artificial Intelligence, Machine Learning, Deep Learning, Computer Science, or a closely related field.

Candidates will be evaluated primarily on the quality, relevance, originality, and depth of their research contributions.

What We Value

We value researchers who can demonstrate the following research capability:

The successful candidate does not need to be an expert in every area of AI.

We are looking for strong researchers with deep expertise in a particular AI/ML domain who can contribute to a broader AI research organization, collaborate across disciplines, and learn adjacent technologies when required.