15 years of human capital research | 32+ companies served | cash-flow positive since 2022 | Seed round, ₹3 Crore
1 / 10
The Problem
Faster EBITDA growth.
Technically strong. Revenues proven. The boat is the company. The net is their network. The fish are their customers. One boat, two nets.
The net they have
Their network. Every buyer in it already knows them, so the ceiling on growth is the size of the network.
The sharks
High margin customers. They will not respond to regular sales methods. Same factory, same cost, more EBITDA per order.
The whales
Export markets. A network does not travel abroad. These are the largest orders they have never quoted.
Margin rises with the size of the customer. Same factory, same cost base, and technical marketing is the only thing between them and it.
2 / 10
The Solution
PDGMS helped a ₹336 Cr yarn maker pitch a concept to a ₹10,000 Cr brand.
The brand named thermoregulation as the axis that interests them. The concept is on their table.
What arriving early is worth, where it is measured
5–15%
extra gross margin paid for creating the demand rather than filling it · semiconductor distribution
25–30%
price premium held in a commodity rope market by selling the application · Garware
5–7 yrs
exclusive supply once the molecule is qualified in · PI Industries
3 / 10
Traction
The workflows our Fellows built by hand now run in the platform.
An environment where the whole problem-solving lifecycle runs. The Fellow does it the right way, and hands it to the organisation.
India’s largest probiotic maker
300+ SQLs
₹4 Cr+
Fellows profiled and mailed, by hand
→ Lead workflow, AI-run
AI AdTech, AI-driven advertising
5-yr skill
in 1 year
Fellows trained freshers, by hand
→ L&D, AI-run, personalised
API maker, NASA-scientist founder
Real-time
by exception
Fellows chased actuals, by hand
→ Plan vs actuals, AI-run
Demand, capability, delivery. One environment runs all three, and the workflow stays after the Fellow leaves.
History of pivots · revenue grew every year
32+ organizations, cash-flow positive since 2022. We did not pick this market. We discovered it from within.
4 / 10
The Market
A ₹325 Cr India beachhead · 1,301 DSIR R&D makers
We start where we already win: ~1,301 DSIR-registered R&D makers under ₹500 Cr, a ₹325 Cr serviceable market our technical-marketing wedge can reach. The same engine opens ~100,000 makers globally ($1.5B). That is the upside, not the headline.
The pool we hold12,000profiled1,100qualified today+2,000each month1,301 is the registry. The call list is ours, and it compounds.
Hidden Champions | They dominate the global supply chain
They already pay for high-quality software and services.
~3,400 · elite
~30,000-50,000 · mid-market
~100,000 · R&D-active mid-caps (the cream)
Out of scope by design
✕ 60,000 MNCs
bureaucratic, multi-year cycles
✕ 358M SMBs
price-led, high churn
TAM · GLOBAL
$1.5B
100k × ~$15K · the upside option
SAM · INDIA BEACHHEAD
₹325 Cr
1,301 makers · what we win first
SOM
₹32 Cr
~130 × ₹25L Yr-1 · India · 3-5 yrs
India entry. 1,994 DSIR-registered R&D centres; 79% under ₹1,500cr, 65% under ₹500cr (1,301 in our SAM). We earn ₹25L in Year 1 (₹10L platform + a ₹15L FDE deployment), and the account holds ₹25L while the FDE expands into the next project. The SAM is computed at ₹25L. See Slide 6.
Why they compound. 3,400+ Hidden Champions, 1,300 in Germany. They grow ~10% a year, 2.5x in a decade, file 5x the patents per head of larger firms, and have minted ~200 billionaires.
Sources. Tier 1: Hermann Simon (~3,400 Hidden Champions). Tier 2: Eurostat SBS 2024 (251,000 EU medium firms), NCMM (US middle market). Tier 3: WIPO IP Indicators 2025 (patent filings), NSF BERD (R&D performers). Base: World Bank (~358M MSMEs). SAM: DSIR (1,994 firms, 65% under ₹500cr). TAM/SOM method: accounts × ACV, SOM as % of SAM (HG Insights, Pear VC). Tier 2-3 firm counts are triangulated estimates from the cited counts.
5 / 10
Go-to-Market
We win the first 50. The platform does the rest.
How we win a client · the 3-day Growth Bootcamp
Leadership arrives at the solution; the FDE codes the prototype live. 50 clients in 12 months.
3,200
connected calls
25%
→
800
profile corrections
50%
→
400
meetings
20%
→
80
bootcamps
25%
→
20
customers
Artifact before ask at every gate, and the artifact is what the previous ask promised. We sell the next artifact, never the engagement. 20 customers per four months. Fifty in twelve is that run rate, held.
The product runs the bootcamp · what every prospect sees in PDGMS
pdgms · their Growth Space
Growth Charter
Growth Workflows
Deliverables
Roadmap Visual
Previews
Proposal
YOUR GROWTH, AS AI WORKFLOWS
today
₹24 Cr
potential
₹32 Cr
AI Scheduling
Demand Forecasting
Shopfloor Quality
Account Intelligence
+ ₹8 Cr EBITDA · every rupee traced to a workflow
Their growth as AI workflows before they sign. The ₹375 Cr yarn maker, just won at ₹2.1L/mo platform setup, no negotiation.
Scale · who delivers as we grow
DT can do about 100 deployments itself; certified partners and self-serve take it to 100,000, using reusable ontology blocks and 15-day training
DT deploysNOW
first ~100 · deep, flagship, high-revenue
Certified agencies
DT-certified partners deploy
DT-Academy students
certified, picked from a portal
Client self-serveAT SCALE
ontology blocks · 15-day setup
In year one, our FDE does the hard part, the change management and the first build, then hands over the keys. As we build more of this into the product, each client needs us less, until a new one can run it themselves from day one.
6 / 10
Business Model
₹25L ACV, held · 44% → the 75% we already earn on HCD
Hidden Champions have the R&D the world wants, but they sell inside their known network. Our FDE studies their PMF, institutionalizes their sales engine, they mint money, we grow with them.
UNIT ECONOMICS
ACV, HELD
₹25L
FDE expands into the next project
AFTER THE FIRST 50
₹10L
platform runs it, FDE exits
CAC
₹2L
at 50 clients
WHAT WE SELL
FDE
deploys the platform, then exits
₹5–20L / yr
20–30% gross margin
PDGMS
the recurring asset · BYOK
₹1–10L / yr
70–75% gross margin
Why ₹25L, not $1M. A formal ontology an AI can operate → reused across clients, not rebuilt → self-served on a visual flow. The platform does the work, not a person.
The first 50 are the training set. We keep the FDE; it moves to the next project inside the account, which holds ₹25L, and writes execution data into PDGMS. We could take the 75% today. We are spending it on the data that lets the platform run without an FDE at all.
# FDE exit: The product promise is that the FDE exits on workflow stabilisation and the account steps down to ₹10L, with all tools and workflows housed in PDGMS. For the first 50 we deliberately do not exit: the FDE is redeployed onto the next project inside the account. Clients may absorb the FDEs directly.
Margins & CAC: DeepThought internal. CAC ₹2L at 50 clients, inclusive of the one-time ₹1.5Cr brand engine (Slide 8) that compounds beyond the first cohort. Growth uplift: IDC / Gartner. Segment: Hidden Champions (Simon; Springer). Reconciles to Slide 5: 50 flagship accounts holding ₹25L ACV on new projects (expansion across the execution stack) = ₹12.5 Cr ARR.
7 / 10
Competition
Mission-critical AI, made feasible · only PDGMS is both
Palantir proved this category, AI that runs the work, is worth a fortune. Serving a ₹375 Cr maker at ₹25L was never its job. That is the maker we serve, with the same kind of execution AI, at a price they can pay.
What PDGMS does not do✗ Factory floor. ✗ Working capital. Two modules are live: CRM (constraints in marketing and sales) and HCD (performance management, daily plan and report, weekly cadence, L&D, execution monitoring by layer). Demand side and people side. That is the beachhead.
MOAT 1 · CHANGE MANAGEMENT, FORMALIZED
Encoded change management, not an interface
Formalized from 15 years of cognition research and 32 builds, so one trained operator runs what used to take a team of engineers and change managers. Today a DT FDE; at scale, a client's own first-principles hire. The operator is interchangeable; the process is the moat.
MOAT 2 · THE ONTOLOGY + ITS COMPOUNDING CORPUS
A formal ontology an AI can finally operate
The science: a formal, computable execution ontology, a 12-stage / 74-node grid from 32 builds and 38 frameworks, academic for thirty years because only a human could operate it. An AI now can. The durability: ship it and the schema can be copied; the corpus of 32 builds compounding it cannot. Each deployment makes the next one sharper.
THE ONTOLOGY, WORKING
That ₹375 Cr maker, read from 3 discovery calls.
Founder's brief: speed up production, delivery runs 13 days against a 10-day promise.
The ontology scored 74 nodes and located the break elsewhere: no one owns improvement (node I3), no way to test a fix (I5), and the real constraint sits upstream in the order book, not on the floor.
→ Build the design layer, not a faster line.
Competitor positions: vendor AI products (Agentforce, Breeze, Now Assist) and Palantir filings (enterprise ACV). Moats: DeepThought HCD-to-FDE pipeline and a 15-year human-capital-development ontology. CRM fit: TruSummit, ARP Ideas.
8 / 10
Team
The founder writes the core. We train the operators.
TarunFounder · mathematician, architect of the core
He owns the abstract architecture and writes the PDGMS core: the execution ontology, the 38 frameworks, the Context OS. 3 engineers complete the build and test cases from his roll-out plans; 10 interns extend it through a sandbox; he runs the whole build AI-assisted.
Mathematics & Theoretical CS · researched with Dr. Vijay Bhatkar (PARAM supercomputer) and Prof. R. Ramanujam (IMSc) · IAS & KVPY research fellow (IIT-H · IISER Pune · IMSc) · 2nd, Madhava Mathematics Olympiad (NBHM, Govt of India) · TEDx speaker.
Operators we manufacturednow embedded at clients, running ₹100–500Cr companies · proof the factory works, not our payroll
Sravan Kumar
Chief of Staff
Pharma · ₹200Cr+
Works directly with the MD, a NASA-scientist founder. Brought in AI that cut purchase-to-dispatch 3 days → 1, and runs decisions that reshape the factory floor.
BBA fresher→Chief of Staff
Shagun Mishra
AI Programs Lead
Probiotics · 60+ countries
Shipped 58 software modules as a 2023 fresher, before AI tools existed. Led a 20-member team; now runs AI Programs at the largest probiotic maker, in 60+ countries.
BTech fresher→AI Programs Lead
Jayaraj
Engagement Lead
DT client companies
Right hand to company Directors on growth strategy and execution, driving organisational transformation across manufacturing and services clients.
BTech fresher→Engagement Lead
Leadership & mentors
Gopala Krishna
President
30+ yrs SME strategy · lead consultant UNDP, World Bank, ADB, GIZ · TISS faculty.
Build team: founder (writes the core) + 3 engineers + 10 interns on a sandbox. The three operators shown are employed by client companies, not DeepThought. They evidence the factory's output. Intake: DeepThought Fellowship (5,000 applications/mo → 25 virtual-tour invites → 5 hires). Reconciles to Slide 8 and Slide 6.
Software can be bought. A founder who writes the core, and the operators we train to run it, cannot be.
9 / 10
The Ask
₹3 Crore to win 50 Hidden Champions and reach ₹12.5 Cr ARR.
Half the round builds demand, because winning the 50 is the only hard part. Delivery pays for itself: each client's ₹25L funds its own FDE. We win them by landing their high-margin customers, then expand into the rest of their execution stack, so the 50 hold ₹25L ACV on new projects, and the cohort holds ₹12.5 Cr ARR. That expansion is what re-prices the next round.
Use of ₹3 Cr
DEMAND ENGINE · ₹1.5 Cr
PR & Events
₹1 Cr
Referral bonuses
₹50L
BUILD, DELIVER & RUN · ₹1.5 Cr
2 elite FDEs
₹40L
Product eng
₹40L
Sales team
₹40L
Misc
₹30L
Delivery self-finances
Each client's ₹25L pays for its own FDE, one FDE serves three. Only the first two are seeded here.
Already cash-flow positive, so the raise buys speed, not survival. CAC settles at ₹2L across the 50, inclusive of the one-time brand engine that compounds beyond the first cohort.
₹60-65L current revenue · cash-flow positive since 2022 · PDGMS live · a ₹375 Cr maker on retainer · 100,000 deployments by 2047 · tarun@dtgrowthteams.com
