Manage the architectural design and deployment of AI solutions, optimize models, develop data pipelines, and lead cross-functional collaboration in AI engineering.
Job Summary:We are seeking a highly skilled AI Engineer to bridge the gap between experimental machine learning models and production-ready intelligent systems. In this role, you will be responsible for the architectural design, optimization, and deployment of Agentic AI solutions and Large Language Models (LLMs/SLMs). You will focus on building robust pipelines, optimizing model performance for specific hardware constraints, and developing sophisticated agent orchestration frameworks to solve complex business automation challenges.Key Responsibilities:•AI System Architecture & Implementation
• Model Deployment & Optimization: Lead the end-to-end integration of machine learning models and fine-tuned SLMs into production environments, focusing on model compression, latency reduction, and hardware-specific optimization.
• Agentic Workflows: Design and implement autonomous agent architectures, including multi-step reasoning engines, tool-use integration, and structured decision-making frameworks.
• Efficient Fine-Tuning Implementation: Develop and maintain the infrastructure for Parameter-Efficient Fine-Tuning (PEFT). Implement techniques like LoRA, QLoRA, or Adapter-tuning to minimize computational overhead.
• Retrieval Augmented Generation (RAG): Build and maintain high-performance vector databases and semantic search indices to enable context-aware AI responses and sub-second data retrieval.
2. Data Engineering & Pipeline Development
• Automated Data Pipelines: Develop scalable, automated pipelines for the cleaning, normalization, and feature engineering of high-velocity raw data streams.
• Quality Assurance: Collaborate with Data Scientists to establish "Ground Truth" datasets and implement automated validation layers to ensure model output reliability and safety.
• System Monitoring: Design and implement monitoring solutions to track model drift, inference performance, and resource utilization in production.
3. Technical Leadership & Integration
• Cross-Functional Collaboration: Work closely with Data Science, Data Engineering, and DevOps teams to ensure seamless transition from model prototype to hardened production binary.
• Mentorship: Provide technical guidance and code reviews for junior engineers, championing best practices in software engineering and AI deployment.
• Stakeholder Engagement: Translate complex technical constraints (e.g., memory limits, inference speed) into clear trade-offs for client stakeholders and project leadership.
Required Skills and Experience:• Experience: Minimum of 5+ years of experience in Software Engineering or Machine Learning Engineering, with a proven track record of deploying AI models in production.
• Technical Stack (Expert Level):
o Languages: Expert proficiency in Python; familiarity with lower-level languages (C++/Rust) or Go for performance-critical components is preferred.
o AI Frameworks: Deep experience with PyTorch, TensorFlow, or JAX, and libraries for model adaptation and inference (e.g., Hugging Face ecosystem).
o Data Infrastructure: Hands-on experience with SQL/NoSQL databases, Vector Databases, and cloud-native AI services (AWS, GCP, or Azure).
• Engineering Rigor: Demonstrated mastery of version control (Git), CI/CD pipelines, containerization (Docker/Kubernetes), and API design (REST/gRPC).
• Problem Solving: Proven ability to optimize models for restricted resource environments (memory, CPU/GPU limits) without compromising core performance
• PEFT & Adaptability: Deep experience with PEFT libraries (e.g., Hugging Face PEFT) and fine-tuning frameworks. Ability to manage and version multiple "Specialist Adapters."
Preferred Qualifications:• Experience: Minimum of 5+ years of experience in Software Engineering or Machine Learning Engineering, with a proven track record of deploying AI models in production.
• Technical Stack (Expert Level):
o Languages: Expert proficiency in Python; familiarity with lower-level languages (C++/Rust) or Go for performance-critical components is preferred.
o AI Frameworks: Deep experience with PyTorch, TensorFlow, or JAX, and libraries for model adaptation and inference (e.g., Hugging Face ecosystem).
o Data Infrastructure: Hands-on experience with SQL/NoSQL databases, Vector Databases, and cloud-native AI services (AWS, GCP, or Azure).
• Engineering Rigor: Demonstrated mastery of version control (Git), CI/CD pipelines, containerization (Docker/Kubernetes), and API design (REST/gRPC).
• Problem Solving: Proven ability to optimize models for restricted resource environments (memory, CPU/GPU limits) without compromising core performance
• PEFT & Adaptability: Deep experience with PEFT libraries (e.g., Hugging Face PEFT) and fine-tuning frameworks. Ability to manage and version multiple "Specialist Adapters."
Education & Shift timingsB.Tech or B.E, in Computer science, software engineering.
• Work Model: Willingness to align with the eClerx’s guidance on WFO-WFH models.
• Shift Timings: Alignment with the group’s work timings (1:00 PM to 10:00 PM IST).
About the Team
• Model Deployment & Optimization: Lead the end-to-end integration of machine learning models and fine-tuned SLMs into production environments, focusing on model compression, latency reduction, and hardware-specific optimization.
• Agentic Workflows: Design and implement autonomous agent architectures, including multi-step reasoning engines, tool-use integration, and structured decision-making frameworks.
• Efficient Fine-Tuning Implementation: Develop and maintain the infrastructure for Parameter-Efficient Fine-Tuning (PEFT). Implement techniques like LoRA, QLoRA, or Adapter-tuning to minimize computational overhead.
• Retrieval Augmented Generation (RAG): Build and maintain high-performance vector databases and semantic search indices to enable context-aware AI responses and sub-second data retrieval.
2. Data Engineering & Pipeline Development
• Automated Data Pipelines: Develop scalable, automated pipelines for the cleaning, normalization, and feature engineering of high-velocity raw data streams.
• Quality Assurance: Collaborate with Data Scientists to establish "Ground Truth" datasets and implement automated validation layers to ensure model output reliability and safety.
• System Monitoring: Design and implement monitoring solutions to track model drift, inference performance, and resource utilization in production.
3. Technical Leadership & Integration
• Cross-Functional Collaboration: Work closely with Data Science, Data Engineering, and DevOps teams to ensure seamless transition from model prototype to hardened production binary.
• Mentorship: Provide technical guidance and code reviews for junior engineers, championing best practices in software engineering and AI deployment.
• Stakeholder Engagement: Translate complex technical constraints (e.g., memory limits, inference speed) into clear trade-offs for client stakeholders and project leadership.
Required Skills and Experience:• Experience: Minimum of 5+ years of experience in Software Engineering or Machine Learning Engineering, with a proven track record of deploying AI models in production.
• Technical Stack (Expert Level):
o Languages: Expert proficiency in Python; familiarity with lower-level languages (C++/Rust) or Go for performance-critical components is preferred.
o AI Frameworks: Deep experience with PyTorch, TensorFlow, or JAX, and libraries for model adaptation and inference (e.g., Hugging Face ecosystem).
o Data Infrastructure: Hands-on experience with SQL/NoSQL databases, Vector Databases, and cloud-native AI services (AWS, GCP, or Azure).
• Engineering Rigor: Demonstrated mastery of version control (Git), CI/CD pipelines, containerization (Docker/Kubernetes), and API design (REST/gRPC).
• Problem Solving: Proven ability to optimize models for restricted resource environments (memory, CPU/GPU limits) without compromising core performance
• PEFT & Adaptability: Deep experience with PEFT libraries (e.g., Hugging Face PEFT) and fine-tuning frameworks. Ability to manage and version multiple "Specialist Adapters."
Preferred Qualifications:• Experience: Minimum of 5+ years of experience in Software Engineering or Machine Learning Engineering, with a proven track record of deploying AI models in production.
• Technical Stack (Expert Level):
o Languages: Expert proficiency in Python; familiarity with lower-level languages (C++/Rust) or Go for performance-critical components is preferred.
o AI Frameworks: Deep experience with PyTorch, TensorFlow, or JAX, and libraries for model adaptation and inference (e.g., Hugging Face ecosystem).
o Data Infrastructure: Hands-on experience with SQL/NoSQL databases, Vector Databases, and cloud-native AI services (AWS, GCP, or Azure).
• Engineering Rigor: Demonstrated mastery of version control (Git), CI/CD pipelines, containerization (Docker/Kubernetes), and API design (REST/gRPC).
• Problem Solving: Proven ability to optimize models for restricted resource environments (memory, CPU/GPU limits) without compromising core performance
• PEFT & Adaptability: Deep experience with PEFT libraries (e.g., Hugging Face PEFT) and fine-tuning frameworks. Ability to manage and version multiple "Specialist Adapters."
Education & Shift timingsB.Tech or B.E, in Computer science, software engineering.
• Work Model: Willingness to align with the eClerx’s guidance on WFO-WFH models.
• Shift Timings: Alignment with the group’s work timings (1:00 PM to 10:00 PM IST).
About the Team
eClerx is a global leader in productized services, bringing together people, technology and domain expertise to amplify business results. Our mission is to set the benchmark for client service and success in our industry. Our vision is to be the innovation partner of choice for technology, data analytics and process management services. Since our inception in 2000, we've partnered with top companies across various industries, including financial services, telecommunications, retail, and high-tech. Our innovative solutions and domain expertise help businesses optimize operations, improve efficiency, and drive growth. With over 18,000 employees worldwide, eClerx is dedicated to delivering excellence through smart automation and data-driven insights. At eClerx, we believe in nurturing talent and providing hands-on experience.
eClerx is an Equal Opportunity Employer. All qualified applicants will receive consideration for employment without regard to race, color, religion, sex, national origin, disability or protected veteran status, or any other legally protected basis, per applicable law.
Top Skills
AWS
Azure
C++
Docker
GCP
Git
Go
Grpc
Hugging Face
Jax
Kubernetes
NoSQL
Python
PyTorch
Rest
Rust
SQL
TensorFlow
Vector Databases
eClerx LLC Mumbai, Maharashtra, IND Office
Express Towers, 4th Floor, Ramnath Goenka Marg, Nariman Point, Mumbai, Maharashtra, India, 400021
Similar Jobs
Analytics
The Associate Process Manager will analyze trade data, manage reconciliation processes, lead stakeholder discussions, and create reports for financial operations.
Top Skills:
ExcelMs WordPowerPointTlm Platform
Analytics
Manage multiple Middle Office and Regulatory support processes, leading teams, implementing improvements, and ensuring accuracy in trade data analysis while fostering client relationships.
Top Skills:
Auditing ProcessesData NormalizationReporting
Analytics
Responsible for managing and developing APIs as part of the process team, ensuring smooth integration and functionality.
Top Skills:
APIs
What you need to know about the Mumbai Tech Scene
From haggling for the best price at Chor Bazaar to the bustle of Crawford Market, the energy of Mumbai's traditional markets is a key part of the city's charm. And while these markets will always have their place, the city also boasts a thriving e-commerce scene, ranking among the largest in the region. Driven by online sales in everything from snacks to licensed sports merchandise to children's apparel, the local industry is worth billions, with companies actively recruiting to meet the demands of continued growth.
