Different sectors, and underneath them the same constraint
A pilot that stalls at the third meeting. A product that turns out to be wrong for the customer you chose. A first channel that works, and then stops working at four times the volume.
I have had all three happen in companies of my own, and paid for them. From 2023 the last one sold into South Africa, the United States, the Middle East and the Philippines, where the argument about regulation that had worked at home did not survive the first meeting. That is the experience I bring, and the whole of it is on the journey.
Each sector, and the business under it
A company that is finished A company that is still running
What I am useful for
-
Strategy
What the company is actually for, which of the five things on the roadmap matters, and what to stop doing. Mostly the second and the third.
-
Market judgement
Whether the market you are in will pay, whether the one beside it would pay more for the same capability, and whether the next country is a market at all or only a map. Cogxio's matching engine turned out to be worth more pointed at an insurer than at an audience. Reading that correctly, and late, is the reason there was a company left to sell.
-
Go-to-market and sales
Getting into regulated buyers, partnerships that survive the pilot, pricing a customer can explain to their own finance team, and what to do when the thing you built is not the thing being bought. In a regulated buyer the meeting that matters is rarely the one you were invited to, and the person who can say yes is usually not in it.
-
Growth
Acquisition and retention, the first channel that genuinely works, what it costs, and the numbers that tell you whether a second one is worth opening.
-
Product and design thinking
Deciding what the product will refuse to do. Most product arguments turn out to be arguments about which customer has been chosen, held in the wrong language.
-
Team building and culture
The habits a team of eight has that a team of eighty inherits whether you like it or not. Setting them on purpose while they are still cheap to set.
-
Hiring
The founding team, the first twenty hires, and what to test for at each stage. Which early hire is quietly deciding the culture, and what you are actually paying for.
-
Brand
A brand a customer will repeat to somebody else, and a founder who can be found and read. Built out of what the company does rather than out of a positioning exercise.
-
AI, applied to a business that already works
Where it earns its place and where it is being bolted on because somebody asked for it. What can be trusted to a machine, what has to stay deterministic, what it costs to run once it is live, and how to keep it honest about the moment it degrades.
-
Fundraising readiness
Whether to raise at all, what the numbers have to look like before the conversation is worth having, and the story and the data room that survive a diligence.
-
Due diligence
Reading a company from its architecture, its team and its books before an investment or an acquisition. I have been on both sides of this, including the one where the company being read was mine.
-
Sector judgement
Insurance and health, devotion and pilgrimage, infrastructure, education, consumer. Each one is set out further down with the business it came out of.
-
Selling outside your home market
An Indian company, taken from 2023 into South Africa, the United States, the Middle East and the Philippines. A product that removes the same piece of paperwork travels. Everything you learned about how your home regulator behaves does not, and the first meeting is where you find that out.
-
Introductions
Insurers, investors, temple and travel operators, infrastructure clients and other founders. I make an introduction when I would put my own name behind both sides of it, and not otherwise.
-
Governance and digital programmes for NGOs
Board and governance basics, a programme that fits the people who will actually run it, and the reporting a funder will read rather than file.
Who I am working with now
-
RODIC Consultants
AI, product and strategy advisor From August 2026Strategy, product and applied AI across highways, bridges, tunnels, metro, roads, power and energy. Advisory work; these are not companies I run.
-
Artivatic
Advisor to the group From September 2026The company I founded, and the group it now sits at the head of. The executive role ended on 31 August 2026 and I advise the group. I do not run it and I do not speak for it.
-
Advisory roles with early-stage startups and NGOs
Product, go-to-market and AI systems for founders; governance and digital programmes for non-profits.
-
Artivatic.ai
Advisor September 2026The company I founded and ran until 31 August 2026. It is now the holding company for the group that trades as InsuranceDekho, and I advise it.
-
RODIC Consultants
AI, product and strategy advisor August 2026Highways, bridges, tunnels, metro, power. Where the work is measured on the ground and paid for on paper.
-
Inspiring Senior Foundation
Advisor September 2026A non-profit working with older people.
For an advisory conversation, write to hi@layaksingh.com or use the contact page, with the stage you are at and the decision in front of you.
If a call is easier, there are fifteen minutes here at no cost. To go straight at one problem for an hour, book a paid session. Come with something specific either way.
The sectors in detail
-
Insurance, health and financial services
Artivatic.aiChanging how insurers, brokers, reinsurers and health partners decide: who to cover, which claim to pay, which one to look at twice. India first, and from 2023 South Africa, the United States, the Middle East and the Philippines. How these buyers procure, why a pilot stalls at the third meeting, what a regulator will and will not accept, and what it takes to get one step of an existing process changed.
Written on this: What 8 Years in Indian Insurtech Taught Me About Where the Money Actually Moves
-
AI, LLMs and agents
Artivatic.ai, DharmikVibes, RODIC ConsultantsMachine learning in production from well before it was fashionable, and agents from having to take commercial decisions about them week by week. It was the product at Artivatic, it is what holds DharmikVibes together, and it is the brief at RODIC. The useful question is rarely which model, but what the business is willing to be wrong about, and how quickly it will find out that it was.
Written on this: Persistent AI Agents: 5 Practical Workflows Worth Testing
-
Infrastructure: highways, bridges, tunnels, metro, roads, power and energy
RODIC ConsultantsProject intelligence and monitoring across highways, bridges, tunnels, metro, roads, power and energy, advising RODIC Consultants and RODIC Digital on strategy, product and AI. Advisory work; these are not companies I run. The client is often a government, the feedback loop runs in years, and the information arrives from a site inspection rather than a browser.
Written on this: Beyond Roads and Bridges: How Bharat’s Infrastructure Story Is Changing
-
Education
iwanttutor, eStudyCornerOnline tutoring and education products, including selling to parents in small Indian cities and to students outside the country.
-
Travel and pilgrimage
DharmikVibesPilgrimage travel: darshan, a bed near the temple and the journey between them. Travel where the calendar sets demand rather than the customer, so load arrives in spikes on dates fixed years in advance, and where nothing sells without trust.
Written on this: The Seeker’s Orbit: How India Will Plan Its Spiritual Yatras in the Age of AI
-
Religion and devotion
DharmikVibesVerifying practitioners, holding a ritual catalogue that is correct across traditions and languages, and settling money to thousands of small partners.
-
Consumer and social
dateIITians, CogxioClosed and verified networks, and the particular economics of a business where the supply is other people: you are selling the confidence to meet a stranger, and growth that costs you that confidence is not growth.
How to start
Write with the problem, where the company is, and the decision in front of you. You will get a straight answer about whether I am the right person, and a name if I am not. I keep the number of engagements low enough that I can hold the whole picture of each one.