Machine Learning Engineer

Machine Learning Engineer Location: Pune

Experience: 2+ years of relevant hands-on experience

Employment Type: Full-Time

Work Mode: On-site

Eligibility: Completed PhD holders only

Please note that a completed PhD is required for this position. Candidates who are currently pursuing a PhD are not eligible.

About Skylark Labs At Skylark Labs, we pioneer embodied artificial intelligence that seamlessly integrates into physical devices and real-world environments, evolving toward true general intelligence. Our mission is to create adaptive AI systems that enhance safety, intelligence, and connectivity across critical real-world applications.

We are building scalable AI-driven visual intelligence systems for real-world deployment across smart cities, manufacturing, surveillance, public safety, and other challenging environments. This role offers an opportunity to work on practical computer vision and deep learning problems where research needs to translate into reliable, high-performance systems.

Role Overview We are looking for a Machine Learning Engineer with a strong combination of research depth and hands-on engineering experience in computer vision and deep learning.

The ideal candidate should have a completed PhD in a relevant field and demonstrated experience taking machine learning models from research or experimentation through optimization, deployment, and real-world evaluation. Candidates should be comfortable working with large-scale visual data, developing custom deep learning solutions, optimizing models for inference, and solving practical computer vision challenges.

We are particularly interested in candidates who can demonstrate personally implemented systems, published research, strong technical projects, or deployed AI solutions , rather than candidates whose experience is primarily academic or theoretical.

Key Responsibilities Design, develop, train, evaluate, and deploy robust computer vision and deep learning models for object detection, classification, segmentation, pose estimation, tracking, OCR, and related visual intelligence tasks.

Develop and fine-tune deep learning architectures using modern frameworks and select appropriate architectures, training strategies, loss functions, and evaluation methodologies based on the problem requirements.

Work with large-scale image and video datasets, including dataset curation, preprocessing, augmentation, annotation strategies, quality analysis, and data pipeline development.

Research and implement approaches to improve model accuracy, robustness, generalization, latency, and computational efficiency.

Apply model optimization techniques such as quantization, pruning, knowledge distillation, and other inference optimization methods for efficient deployment.

Optimize machine learning inference for edge and cloud environments, including systems with hardware acceleration and resource constraints.

Integrate computer vision and machine learning models into production pipelines and real-time video processing systems.

Work with camera and video-streaming systems and develop solutions capable of operating under real-world conditions.

Collaborate with MLOps, DevOps, product, and engineering teams to take models from experimentation to reliable production deployment.

Investigate model performance issues, analyze failure cases, and develop data-driven improvements to system accuracy and reliability.

Stay current with relevant research and emerging techniques in computer vision, deep learning, multimodal AI, and visual intelligence, and evaluate their practical applicability to Skylark Labs' products and systems.

Required Skills & Qualifications Candidates must have a completed PhD in Computer Science, Artificial Intelligence, Machine Learning, Computer Vision, or a closely related technical field.

Candidates should have 2+ years of relevant hands-on experience in machine learning, computer vision, deep learning, or related engineering/research roles, excluding coursework and short-term internships.

Strong proficiency in Python and experience with modern machine learning and deep learning frameworks are required.

Strong understanding of computer vision and deep learning fundamentals, including model architecture, training, evaluation, optimization, and performance analysis, is expected.

Candidates should have hands-on experience developing and training custom machine learning or computer vision models rather than only using pre-trained models as black-box solutions.

Experience working with large-scale image and video datasets and building reliable data preparation or processing pipelines is required.

Experience deploying and optimizing machine learning models on edge computing devices or hardware-accelerated platforms is highly relevant to this role.

Strong understanding of model optimization techniques and practical inference-performance considerations is expected.

Experience with real-time video streams, camera-based systems, or production computer vision applications is preferred.

Proficiency with Git or other version control systems and experience working in collaborative engineering environments is expected.

Research & Technical Depth A strong research background is highly valued for this role. Candidates with peer-reviewed publications, patents, significant research projects, open-source contributions, or demonstrated technical work in computer vision and machine learning are encouraged to apply.

Candidates should be able to clearly explain the problem they solved, the approach they developed, the experiments they conducted, the technologies they used, and their personal contribution to the work.

Experience translating research ideas into working prototypes, optimized models, or production-ready systems will be particularly valuable.

Nice-to-Have Skills Experience with model serving frameworks and inference optimization tools.

Experience with MLOps practices, CI/CD pipelines, experiment tracking, model versioning, or automated deployment workflows.

Knowledge or hands-on experience with Vision-Language Models (VLMs), multimodal AI, or other advanced visual understanding systems.

Experience with cloud AI services and edge AI platforms.

Experience with real-time object tracking and multi-object tracking systems.

Experience with OCR and document/image understanding systems.

Experience with smart surveillance, public safety, industrial vision, or other real-world computer vision applications.

Experience deploying AI models on GPUs or other hardware accelerators.

Experience working with constrained compute, memory, latency, or power requirements in real-world deployment environments.

What We Look For We value candidates who combine strong research fundamentals with practical engineering ability .

Candidates should be able to demonstrate ownership of meaningful technical work and explain their individual contribution rather than only describing work performed by a larger team.

During the interview process, candidates may be asked to discuss their research, publications, technical projects, model architectures, experiments, deployment experience, optimization techniques, and real-world system challenges.

Important Eligibility Requirement This position is open only to candidates who have completed a PhD in Computer Science, Artificial Intelligence, Machine Learning, Computer Vision, or a closely related field.

Candidates currently pursuing a PhD are not eligible to apply.

Candidates should also have relevant hands-on experience beyond academic coursework and short-term internships.

Application & Screening Process Interested candidates are required to complete the initial screening form:

Screening Form:

https://docs.google.com/forms/d/e/1FAIpQLSfK6pmi1a0TAy1MEWp0M4xaS3Aui-OUPLSutXIvlZKUSMYIGA/viewform?usp=sharing&ouid=118009209953067933509

The initial screening includes questions related to the candidate's education, relevant experience, technical background, compensation expectations, hands-on computer vision experience, and evidence of previous technical work.

Candidates should provide accurate and up-to-date information and an accessible résumé/CV link. Where applicable, candidates are encouraged to provide verifiable evidence of their technical work, including research publications, GitHub repositories, patents, technical projects, deployed systems, or other relevant work.

Shortlisted candidates will be contacted for the next stages of the hiring process. During the interview process, candidates may be asked to explain and verify the technical work, research, projects, and experience mentioned in their application.

Please apply only if you meet the completed PhD and relevant hands-on experience requirements .

Incomplete or inaccurate applications may not be considered for further evaluation.

Seniority level: Entry level

Employment type: Full-time

Industries: Robotics Engineering