Product Manager

About the Role

We are looking for an AI Product Manager who combines strong product management fundamentals with genuine hands-on technical expertise in building and deploying AI/ML systems, including Large Language Models (LLMs) etc. This is a hybrid role suited for someone who can move fluidly between writing PRDs and prototyping a model, between running sprint planning and reviewing training data quality. You'll own the product vision, roadmap, and delivery for AI-driven features while staying technically credible with the engineering and data science teams you work alongside.

 

Key Responsibilities

Product Management

• Define and own the product vision, strategy, and roadmap for AI/ML-powered products and features.
• Translate business problems and customer needs into clear product requirements, user stories, and acceptance criteria.
• Manage the end-to-end product lifecycle: discovery, prioritization, development, launch, and post-launch iteration.
• Run agile ceremonies (sprint planning, backlog grooming, retrospectives) and manage execution using Jira, Confluence, and similar tools.
• Balance and manage multiple concurrent product workstreams/portfolios, prioritizing across competing stakeholder demands.
• Define and track success metrics/KPIs (model performance, adoption, latency, cost-per-inference, business impact).
• Partner cross-functionally with engineering, data science, design, legal/compliance, and business stakeholders.
• Communicate roadmap, progress, and trade-offs clearly to leadership and stakeholders.

Hands-On Technical / AI-ML

• Directly contribute to prototyping, fine-tuning, evaluating, and iterating on LLMs and other ML models (e.g., prompt engineering, RAG pipelines, fine-tuning, embeddings).
• Collaborate closely with engineers/data scientists on data pipelines, model architecture trade-offs, and deployment strategy (MLOps).
• Evaluate build-vs-buy decisions for foundation models, vector databases, and AI infrastructure tooling.
• Stay current on the AI/LLM landscape (new model releases, techniques, regulatory/ethical considerations) and translate emerging capabilities into product opportunities.

Qualifications & Experience

• Bachelor of Engineering (B.E.) or B.Tech — Computer Science, Information Technology, Electronics, or related discipline.
• 5 to 7 years of total professional experience, including a 3 years in Product Management (AI/ML or data-driven products preferred).
• Demonstrated hands-on experience building, fine-tuning, or deploying LLMs — not just managing engineers who do so (e.g., portfolio, GitHub, published work, or specific project examples).
• Practical knowledge of Python and common ML/AI frameworks (PyTorch, TensorFlow, Hugging Face) or LLM tooling (LangChain, LlamaIndex, vector DBs).
• Proven experience managing multiple products/workstreams simultaneously in a fast-paced environment.
• Proficiency with product/project management and collaboration tools: Jira, Confluence, or equivalents.
• Strong grasp of agile/scrum methodologies and experience running cross-functional agile teams.
• Excellent stakeholder management, written, and verbal communication skills — able to translate technical concepts for non-technical audiences and vice versa.

Preferred Qualifications

• Domain experience in automotive, mobility, or manufacturing industries is an added advantage.
• Solid understanding of NLP concepts, model evaluation metrics, prompt engineering, and RAG architectures is an added advantage.
• Familiarity with responsible AI practices — bias mitigation, model governance, data privacy, and compliance frameworks.
• Experience with cloud AI platforms (Azure ML, AWS SageMaker, GCP Vertex AI).
• Certifications in Product Management (CSPO, PMP) and/or AI/ML certifications are an added advantage.

What Success Looks Like in This Role

• Shipped AI/LLM-powered features that measurably improve user outcomes and business KPIs.
• A well-managed, transparent roadmap and backlog across multiple concurrent initiatives.
• Strong trust from both business stakeholders (for product judgment) and technical teams (for technical credibility).
• Continuous improvement of model quality metrics through direct hands-on involvement, not just delegation.