Frame the problem
Start with the user’s decision or task, its constraints and the business outcome. Choose AI only when its capabilities fit the problem.
AI Product Manager in Kolkata — Prasanjit Saha
I’m Prasanjit Saha, a Kolkata-based product and digital transformation leader with 16+ years across ecommerce, B2B ecosystems and enterprise platforms. My AI product-management work connects business problems, user workflows, product strategy and hands-on public experiments. Dharma AI, SCORE-AI and PujoPlan AI provide concrete examples of retrieval, structured decision support and AI-assisted operational planning.
What AI product management means to me
AI product management is not just prompt engineering or adding a model endpoint to an existing product. The product work sits in problem selection, workflow redesign, retrieval, evaluation, trust, user experience, economics and adoption.
Design the workflow
Connect the AI interaction to data, systems, human review and the next action. Make uncertainty and failure understandable in the UX.
Evaluate and earn trust
Assess usefulness, grounding, response quality, latency and cost. Learn from real usage and adoption, not only a convincing demo.
Applied AI work
Public experiments make product choices, technical concepts and their limits tangible.
Live
Dharma AI
A conventional chatbot can produce fluent advice without showing what informed it, and a source-heavy experience can be slow or difficult to use. The product challenge was to make a reflective conversation feel simple while preserving source context, scope boundaries and a dependable response experience.
Explore the project →
Live
SCORE-AI
Claims such as “AI-ready” or “AI expert” are easy to make and hard to interpret. A generic score hides the fact that an engineer, product manager, transformation leader and business head may need very different combinations of capabilities. The product challenge was to provide a structured, useful signal without presenting it as a certification, hiring recommendation or prediction of job performance.
Explore the project →Public experiment
PujoPlan AI
Operations teams must maximize useful output within finite time, capacity and resource constraints. Logistics routes, production schedules, sales beat plans and field-service assignments all depend on mandatory tasks, sequence, geography and changing priorities. Powerful planning systems can still be difficult for people to use. PujoPlan AI explores whether natural-language interaction can simplify these decisions without letting a language model invent operational feasibility.
Explore the project →Enterprise product experience
My enterprise experience spans B2B and B2B2C platforms, partner ecosystems, CRM/ERP-connected workflows and banking/payment integrations. The work combines enterprise communication, sales-force automation, multilingual experiences and adoption at scale.
At Berger Paints, that included digital commerce, dealer onboarding, lead-management systems and direct-to-bank incentives. More recent enterprise product work connects sales, CRM and distributor information with field workflows.
Building a Digital Ecosystem for a Large Distributed Influencer Network
How we moved from fragmented offline relationships to a connected digital ecosystem spanning onboarding, engagement, incentives, transactions and repeat participation.
Making Incentives Work in Real Time
How a slow, opaque incentive process was redesigned into a faster and more transparent digital experience—from validation to settlement.
Connecting the Data Behind Everyday Sales Decisions
A case study in redesigning fragmented sales workflows into a connected digital operating system for field teams and managers.
Why enterprise transformation matters for AI
An AI model alone does not make a useful enterprise product. The surrounding workflow often determines whether it creates lasting value: data, approvals, incentives, accountability, integrations and adoption. My background in enterprise transformation is what I bring into AI product thinking.
Explore the broader expertise →Technical fluency, with a product lens
The public AI Lab work provides practical exposure to retrieval-augmented generation, source grounding, structured outputs and scoring. Building and reviewing those experiences raises product questions about response quality, evaluation, latency, trust and interpretation.
Firebase, APIs, cloud deployment, analytics and prototyping help connect a product hypothesis with a working experience. This is technical fluency in support of product leadership, rather than a claim of specialist engineering depth.
Product leadership
Leading a multi-product portfolio means making roadmap trade-offs, aligning commercial and operational stakeholders, and collaborating with engineering and design. Adoption, release governance and measurable business outcomes matter alongside delivery.
Explore my career and leadership approach →Based in Kolkata. Working on problems that are not location-bound.
I’m based in Kolkata, India, and work across enterprise product, digital transformation and applied AI. I’m particularly interested in contributing to the growing product and AI ecosystem in Kolkata while remaining open to national and global leadership, collaboration and speaking opportunities.
Speaking & guest lectures
- AI for Business Leaders →
Turning AI concepts into useful business choices.
- From Digital Transformation to AI Transformation →
Understanding how workflows and organisations change.
- Product Management Beyond Features →
Connecting product decisions with adoption and outcomes.
- Building AI Products Without Being an AI Engineer →
Learning emerging technology by building useful experiments.
- Enterprise Product Management →
Working with commercial complexity, platforms and enterprise workflows.
- The Future of the Product Manager →
Exploring how AI changes the practice of product management.
Perspectives on product & AI
The Next Level of Abstraction: Why AI-Assisted Programming Is a Natural Evolution… Even If It Feels Like Cheating
AI-assisted programming belongs in computing’s long history of abstraction: complexity moves, while human attention moves closer to intent, system design and judgement.
Rethinking Professional AI Readiness: A Role-Relative Framework for Capability, Evidence and Assessment Confidence
Why AI readiness must be interpreted against the responsibilities of a role, with capability, evidence and assessment confidence treated as separate signals.
AI Isn’t Replacing Product Managers — It’s Exposing Who Was Strategic All Along
AI compresses execution-focused product work and raises the value of problem framing, judgement, prioritisation and builder-oriented product leadership.
Frequently asked questions
Who is Prasanjit Saha?
Prasanjit Saha is a Kolkata-based product and digital transformation leader with 16+ years across entrepreneurship, ecommerce, B2B ecosystems and enterprise platforms. His current focus includes AI product management and applied AI.
Is Prasanjit Saha an AI Product Manager in Kolkata?
Prasanjit Saha is based in Kolkata, India, and applies product-management experience to AI products and enterprise workflows. His professional title is Product and Digital Transformation Leader. His AI product-management focus is evidenced by public projects including Dharma AI, SCORE-AI and PujoPlan AI, documented in the AI Lab.
What AI products has Prasanjit built?
Prasanjit’s public experiments explore grounded knowledge experiences, structured decision support and constrained planning. Each project page documents the business problem, product approach and limits; these experiments are distinct from his enterprise career work. Dharma AI, SCORE-AI, PujoPlan AI
What type of AI product work does Prasanjit focus on?
Prasanjit focuses on business problem selection, workflow design, structured AI outputs, retrieval and grounding, deterministic guardrails, evaluation and adoption. PujoPlan AI explores AI interpreting intent while deterministic optimization protects feasibility. The current published AI Lab taxonomy includes RAG, AI orchestration, Grounded responses, Responsible AI UX, Structured scoring, Role context, AI evaluation, Responsible assessment, Operational AI, Constraint Optimization, Geospatial Planning, Human-in-the-Loop AI.
Does Prasanjit work on enterprise AI?
His enterprise background includes B2B platforms, CRM/ERP-connected workflows, banking integrations, incentives, lead management and sales-force automation. His public operational-AI experiment, PujoPlan AI, explores a pattern relevant to production planning, transport management, sales beat plans/PJP and field-service scheduling; it does not claim deployed enterprise use or quantified enterprise results.
Is Prasanjit available for guest lectures or speaking in Kolkata?
Prasanjit welcomes enquiries about guest lectures and speaking in Kolkata and beyond on AI for business leaders, AI product management, enterprise transformation and building practical AI products. Visit the speaking page for topics or use the contact form to discuss a session.
- Name
- Prasanjit Saha
- Location
- Kolkata, India
- Experience
- 16+ years
- Primary focus
- Product Leadership · Enterprise Transformation · Applied AI
- Experience areas
- Ecommerce · B2B/B2B2C platforms · Enterprise systems · Digital marketing · Marketplaces · Fintech · AI products
Bring a real product question.
For a leadership opportunity, applied AI collaboration, enterprise challenge or guest session.
Start a conversation →