I’m Dipankar Sarkar. I run advisory sessions, workshops, and programmes for boards, executive teams, investment portfolios, and institutions. Each is built around a decision or learning outcome, and draws on 18 years of building production AI and software systems.
Current question
What would have to be true before this AI decision deserves a yes?
Each route starts from who must decide or learn, what they must leave with, and what evidence the decision needs.
“Our board must approve an AI initiative. Who can explain the technical choices and leave us with a clear decision and ownership structure?”
You leave with: A board decision session and a written decision record with owners, conditions, and a review date.
“Our executives need enough understanding to make decisions without becoming engineers. What should the session cover and leave behind?”
You leave with: An executive workshop tied to named decisions, with a written decision and action record afterwards.
“Our portfolio companies keep asking for AI help. Can a programme triage who needs education, validation, engineering, or no intervention?”
You leave with: A multi-company AI clinic or capability programme with a triage outcome for every participating company.
Portfolio programmes“Our association wants an AI programme useful to members with different capabilities. How can we structure a practical cohort?”
You leave with: An institutional learning programme with audience objectives, a delivery model, accessibility provisions, and assessed outputs.
Conference talks, panels, webinars, podcasts, and event workshops, without unsupported forecasts.
I am the founder and principal consultant at Neul Labs, and have spent 18+ years building production software, machine-learning products, robotics, and AI agent systems. That work shapes which questions I ask a board, and which evidence I treat as enough.
Evidence
Architected enterprise banking AI with conduct-risk workflows, evidence generation, human review, escalation, evaluation, versioning, and release controls. Part of the Aveni team in the first FCA Supercharged Sandbox cohort; the FCA lists Aveni as an accepted firm, which is not an FCA endorsement or certification.
Inspect the source ↗Kwippy, Jaja.tv, Octo.ai, ExpressMOJO, Boom, and Neul Labs, spanning social software, interactive media, ML and analytics, logistics, blockchain infrastructure, and applied AI. Advisory and investment records are not counted as founded companies.
Read more →Built or led NLP, recommendation, computer-vision, and trust-and-safety ML systems serving more than 100M users. Company-level outcomes are not attributed to one person.
Fellow and Joint Program Leader of the New Delhi chapter: organising cohort sessions, bringing practitioners into the room, and helping founders examine decisions.
Read more →Two Packt books span web infrastructure and everyday AI workflows. What Generative AI extends the teaching into open GenAI and Agentic AI playbooks for business and technical leaders.
7 AI Workflows to Save Hours at Work Every Week
Connected workflows for email, meetings, research, reports, task tracking, spreadsheets, and professional communication.
Read more →More than 100 recipes for Nginx infrastructure
Configuration, application serving, reverse proxies, load balancing, caching, monitoring, and security for Nginx.
Read more →21 open chapters in ten languages
A business-leader guide to GenAI adoption followed by an engineering and operating guide to agent systems.
Read more →Different decisions need different engagements, so each lives on the site built for it.
Contract AI engineering, delivery, and team enablement.
dipankar.co →Hands-on CTO, Head of AI, and interim leadership mandates.
dipankar.name →Research, validation studies, publications, and technical guides.
dipankar.cc →A useful first message
Name the audience, the decision or learning outcome, the date, and the format. I reply personally, and say plainly when the work belongs elsewhere.