New startup ideas · AI and software · AI infra and compute
startup idea
Sovergrid
The procurement marketplace where governments buy sovereign AI compute and models.
Sovergrid is a marketplace where ministries, national labs, and public universities buy accredited AI compute, models, and deployment tooling from vetted domestic and allied providers under pre-negotiated framework contracts.
- Marketplace
- Government and public sector
- $10-100B market
- Platform others build on
- Global from day one
3/5
venture judge
1
similar startups, last 2 years (13 all-time)
99%
of 2 nearest real companies still alive
yes
8 matching federal grants and programs
Direction supported by government programs and grants
Test it before you build it
$1,000 · 6 weeks · 20 prospects
For $1,000 and 6 weeks, prove that US public research buyers with fresh AI compute budgets will commit real dollars in writing through an intermediary inside one quarter - not the 18-36 months the card says will kill the runway.
Riskiest assumption · A US research computing director with 2026-27 AI compute budget can and will commit in writing to purchase through a third-party marketplace via a contract vehicle their institution already holds within one quarter, rather than the 18-36 month default procurement path that outlasts the company's runway.
1Focus group: who and where
The research computing director or CIO at a US national lab, NSF-funded computing center, or public R1 university whose institution has new AI compute money - a Genesis Mission response, an FY27 budget line - and who is under pressure to show committed spend before the fiscal year closes.
where to find 20 · The CASC (Coalition for Academic Scientific Computation) member directory of roughly 90 US research computing centers; the EDUCAUSE Research Computing and Data Community Group; speaker rosters from the PEARC and SC supercomputing conferences; and SAM.gov activity around the Genesis Mission solicitation closing 2026-12-17.
2Sell first, build later
A framework catalog of 5 vetted US compute and model providers at pre-negotiated rates, purchasable through a contract vehicle the buyer's institution already holds, with the first purchase order processable within 60 days of a signed LOI.
the ask · No fee to buyers; providers pay a 6% success fee on settled purchases - so the money-side proof lives on the supply side.
a real yes · A real yes is a buyer LOI carrying a dollar amount, a named vehicle and a dated quarter, paired with a provider-signed 6% agreement. 'Send us more information,' an unsigned interest form, or a conference-hallway commitment is not.
3Small experiments
The first one attacks the riskiest assumption; each ends with a number that says whether to run the next.
1. Budget and vehicle discovery
$300 · 14 days
Build a list of 30 research computing directors from the CASC directory and PEARC speaker rosters, email 20, and run 15 discovery calls asking three questions: what FY27 AI compute money exists, which contract vehicle they already hold could spend it, and how fast a purchase through a third party could clear their procurement office. Log every named vehicle and dollar figure in one sheet.
keep going if · 8 of 15 directors name both a live budget of $250k+ and an existing vehicle that could clear a third-party purchase in under 6 months
2. Provider success-fee agreements
$300 · 14 days
Pitch BD leads at 10 US compute and model providers - Computable from the card plus commercial GPU clouds such as Lambda and CoreWeave - on a non-exclusive referral agreement: 6% success fee on any public-sector purchase Sovergrid originates, no listing cost. A signed agreement is the count; verbal interest is not.
keep going if · 4 of 10 providers sign the 6% referral agreement and supply an indicative public-sector rate sheet
3. Framework catalog LOI test
$400 · 21 days
Assemble the signed providers' rate sheets into a one-page framework catalog and send it to every director who named a budget, with a fill-in letter of intent: dollar amount, named contract vehicle, purchase quarter. Follow up by phone until each says yes or no.
keep going if · 3 signed LOIs totaling $500k+ with a purchase quarter no later than Q2 2027
4Collect a deposit up front
Tesla took $1,000 refundable reservations for the Model 3 and $100 for the Cybertruck before building either: the deposit is the measurement, not the revenue.
$0
no cash yet: take a signed commitment
how · No money changes hands with a public buyer yet - the instrument is a signed letter of intent stating a dollar amount, a named existing contract vehicle, and a purchase quarter, signed by the research computing director or their procurement officer; on the supply side, providers sign the 6% success-fee referral agreement, which is where money-backed commitment is actually collectable now.
what it reserves · A slot in the first buyer cohort, onboarding priority, and the published rate sheet locked for 12 months.
refund · The LOI is non-binding and withdrawable in writing at any time; nothing is owed by the buyer until a purchase order settles.
target · 3 buyer LOIs totaling $500k+ and 4 signed provider agreements within 42 days.
before taking money · This is government procurement: take no fee on any public purchase and sign no agency paperwork until procurement counsel confirms the 6% referral structure complies with federal and state procurement and anti-kickback rules.
Go: build it if
8 of 15 directors name a live budget and a sub-6-month vehicle, 3 LOIs totaling $500k+ are signed with quarters attached, and 4 of 10 providers sign the 6% agreement - the fast procurement path exists, build settlement and compliance mapping around it.
Kill: stop if
Fewer than 3 of 15 directors can name both a budget line and a vehicle faster than 12 months, or 0 signed LOIs after 15 conversations - the 18-36 month cycle is real, and the marketplace dies on runway exactly as the judges said.
5 Scripts to run itoutreach message, landing copy, deposit terms · click to open
outreach message
Your center likely has new AI compute money - Genesis Mission responses, FY27 lines - and the default path to spending it takes 18 months. I'm assembling a catalog of vetted US GPU and model providers at pre-negotiated rates, purchasable through contract vehicles you already hold. Before building anything, I'm asking 15 research computing directors what would make that usable this fiscal year. Can I get 20 minutes with you this week?
landing page
Buy vetted US AI compute through vehicles you already hold No buyer fees; 5 providers at pre-negotiated rates, PO-ready in 60 days. Request the catalog and rate sheet.
deposit terms
No payment is due now. This letter of intent records your center's intent to purchase up to the stated amount of AI compute through the named contract vehicle in the stated quarter; it is non-binding and withdrawable in writing at any time. Signing reserves your center's slot in the first buyer cohort and locks the published rate sheet for 12 months.
Would you run this test?
One tap. The yes-share feeds the Demand pillar of this idea's score; nobody sees who answered.
Budgets are out-of-pocket estimates for a team of one to three, US market. Size the deposit to the deal, and check the terms before taking money in a regulated line.
Scorecard
One score that balances how trendy the idea is, the demand for it and its potential for 100x, with competition measured relative to every other idea in the catalog. Recent startup trends first, government priorities second.
88
Idea Score, 0-100 · raw 54.8 x 1.61
Warm
competition: more crowded than 20% of ideas · headwind x0.90
+4.4
government priorities, secondary (82 matching grants)
Trend
59
Is the wave forming now? 2025-26 entrants vs 2023-24, rounds since 2025, the sector's live-batch direction, the 2026 trend analyst.
- Entrants 2025-26 vs 2023-24 (similar companies)3
- Rounds announced 2025+ in the sector82
- Sector direction (live batch)100
- 2026 trend analyst50
Demand
52
Does anyone want it? YC's current RFS, companies already paid for something similar, the operator judge, founders' yes-rate in decks, readers who would run the test.
- YC asks for it (current RFS: idea / sector)30
- Someone already pays (similar companies, recent / all-time)100
- Operator judge: real pain25
100x potential
58
Can it return a fund? The venture judge (double weight), market-size and moat axes, neighbours still alive, the technologist judge.
- Venture judge50
- Market size axis67
- Moat axis100
- Neighbours still alive81
- Technologist judge0
Score = 100 x cbrt(Trend x Demand x 100x) x (1 - 0.5 x crowding) + government bonus (max 5), calibrated so the 95th-percentile idea scores 90 (order never changes). A geometric mean: a weak pillar cannot be papered over. Percentiles are among the 272 ideas in the catalog; the terms matched were procurement, marketplace, governments, buy, sovereign, compute, models, ministries.
The idea in full
- What
- Sovergrid is a marketplace where ministries, national labs, and public universities buy accredited AI compute, models, and deployment tooling from vetted domestic and allied providers under pre-negotiated framework contracts. Providers and independent software vendors list once and sell across many jurisdictions; the platform handles compliance mapping, procurement paperwork, and usage-based settlement through integrations rather than custom infrastructure. In-country compliance and onboarding teams do the ops-heavy work that makes each new jurisdiction sellable.
- Why now
- Velatir just raised €5M to expand AI infrastructure across Europe, and the Genesis Mission solicitation from the Office of Science (closes 2026-12-17) is pushing US federal AI money toward procurement channels - public AI compute budgets are being created faster than agencies know how to spend them.
- Wedge: first customer and entry point
- One founder-sold framework agreement with a single national research agency, then the universities and labs under that framework as repeat buyers who pull providers onto the platform.
- Path to 100x
- Public-sector AI spend is a $10-100B flow, and every framework contract adds buyers that attract more providers and ISVs, while compliance work done once resells across jurisdictions at software margins. If Sovergrid becomes the default purchasing channel in even 20 OECD countries, a single-digit take rate on that flow with standard winner-take-most marketplace dynamics supports a $1B+ outcome.
- Ceiling
- Sovereignty politics can force one-country forks, fragmenting the network into local marketplaces too small to compound.
- Closest real companies, as the generator saw them
- Velatir builds and operates infrastructure itself rather than running a neutral marketplace; Computable and Touchmark trade raw GPU hours for commercial buyers with no accreditation layer. None lets third parties sell to governments through one compliance and settlement platform.
- Main risk
- Public procurement cycles of 18-36 months outlast the company's runway before the marketplace flywheel starts turning.
Five judges
Each judge scores every idea in the catalog with a named rubric; the venture judge decides whether a card is shown at all (4-5 is venture-grade).
Venture investor
3/5
Single-digit take on $10-100B public AI spend is real, but 18-36 month procurement and in-country ops teams make growth linear.
Bootstrapper
1/5
The card admits 18-36 month public procurement cycles outlast runway, and a marketplace needs both sides before any take rate.
Operator
2/5
Two-sided marketplace against 18-36 month public procurement cycles means nobody's personal pain gets solved before the runway ends.
Technologist
1/5
Self-described integrations and no-code over procurement paperwork means there is no technical wedge a systems integrator could not copy.
Risk
3/5
Framework contracts across many jurisdictions diversify buyers, but 18-36 month public procurement cycles and sovereignty forks are both outside the company's control.
trends
3/5
Genesis Mission closing 2026-12-17 and newly formed sovereign budgets are dated anchors, but 18-36 month procurement cycles outrun the open window.
Similar startups in the directory
Companies whose pitch matches most of the idea's terms (procurement, marketplace, governments, buy, sovereign, compute, models, ministries): 13 all-time, 1 from the last two years. Same matching as Idea Check.
The autonomous back-of-house for restaurants.
Capitol: the sovereign operating system for intelligence.
AI-powered pre-construction platform leveraging Large Language Models to draft and review contracts and RFPs.
The sovereign agentic workstation for molecular discovery
Operator of a B2B marketplace for remote workers intended to provide companies with a comprehensive solution for remote and hybrid work. The company allows companies and individuals to easily find or rent out available office spaces and have the option to give their employees access to book workspaces in any part of the world on their behalf, in a buy now pay later business model, enabling clients to manage their remote workforce efficiently.
Marketplace for vacation home co-ownership, fully managed
Making compute for AI as accessible as oxygen for humans.
Building the NASDAQ for Real Estate.
Digital Co-op Renters Insurance = half the price, better coverage.
Operator of a motorcycle marketplace intended to transform motorcycle sales in Southeast Asia. The company's marketplace offers a wide selection of motorcycles from a variety of brands and models and also provides a search engine that makes it easy to find motorcycles by make, model and location, enabling motorbike owners and two-wheeler enthusiasts to have a comprehensive motorcycle database that provides information about the same and find the suitable motorcycle for their needs.
"Name Your Price" for products
ImaliPay is a financial health platform offering financial products to freelancers and blue collar gig workers.
The generator's reference companies
Real companies the model named as closest when it wrote the card, with their fate. A check mark is a company the radar could verify in its directory.
Public money in this direction
US federal grants, SBIR/STTR awards and open opportunities from the radar's public-money feed, matched to the idea's terms; the sector totals give the context.
8
grants and programs matching the idea
243
startup-relevant grants in AI infra and compute
$339M
awarded in the sector, tracked
8
opportunities open now in the sector
- Novel preservative-free eyedrop with antimicrobial and anti-inflammatory properties for treating Acute Infectious Conjunctivitis and Epidemic KeratoConjuntivitis (EKC)awardhigh relevance
NIH / NEI · SBIR phase I · $307K · posted 2026-09-15
- I-Corps: Translation Potential of Privacy-Preserving Internet of Things (IoT) for Clinical and Home Health Monitoringawardhigh relevance
National Science Foundation · I-Corps · $50K · posted 2026-08-20
- I-Corps: Translation Potential of an Artificial Intelligence (AI) Co-Scientist Platform for messenger Ribonucleic Acid (mRNA) Therapeutics Designawardhigh relevance
National Science Foundation · I-Corps · $50K · posted 2026-08-19
- I-Corps: Translation Potential of an Artificial Intelligence (AI) Surrogate for Coastal Storm Surge and Water Level Forecastingawardhigh relevance
National Science Foundation · I-Corps · $50K · posted 2026-08-19
- GOALI: Machine Learning Enabled Multiscale Modeling Platform and Virtual Sensing Digital Twin for Fatigue in Metals Additive Manufacturing (MAM)awardhigh relevance
National Science Foundation · Mechanics of Materials and Str, AM-Advanced Manufacturing, Special Initiatives · $900K · posted 2026-08-18
- Mid-scale RI-1 (M1: DA): Designing Scalable Community-driven Infrastructure to Support Biodiversity Genomics (SCIBG).awardmedium relevance
National Science Foundation · Cross-BIO Activities, Mid-scale RI - Track 1 · $12M · posted 2026-09-14
- Research Infrastructure: Mid-scale RI-1 (M1:IP) Genome Infrastructure catalyzing next Generation Antarctic Data, Access, Tools and Analyticsawardmedium relevance
National Science Foundation · Mid-scale RI - Track 1 · $18M · posted 2026-09-14
- Collaborative Research: AI-Enabled Digital Twins for Resilient Freshwater Infrastructure and Communities under Cascading Hazardsawardmedium relevance
National Science Foundation · ISP-Infrastructre Systms & Ppl · $122K · posted 2026-09-09
Market signal
What the radar sees in AI infra and compute: new companies by cohort year, the forming YC batch, and outcomes since the February snapshot.
AI infra and compute · 11 → 34 → 59 → 58 → 72 new companies 2022 → 2026 · 95% aliveYC F26 live: 6 in this cluster, 5% of the batch (was 8% in S26)Since February, of 116 YC companies here: 3 acquired, 3 shut down, 26 rewrote their pitch
Design attributes
The card is one cell of a designed set: every axis below was chosen before the text was written, and the text had to realize it.
- Buyer
- Government and public sector
- Business model
- Marketplace
- Path to 100x
- Platform others build on
- Market size
- $10-100B market
- Capital intensity
- Capital-medium (ops, field teams)
- Speed to revenue
- Revenue in 1-3 years
- Technical depth
- Integrations, no-code
- Go-to-market
- Founder-led sales
- Moat
- Network effects
- Geography
- Global from day one
- Regulation
- Some regulation
- Vibe
- Hot space
Listed under
An idea sits in its own sector and in any sector its text clearly touches.
More ideas like this
AI and software · AI infra and compute
Residua
Secondary market where enterprises resell unused committed cloud and GPU spend.
Residua is a self-serve exchange where enterprises list the unused portion of multi-year committed spend contracts with clouds and GPU neoclouds, and other enterprises buy that entitlement at a discount.
AI and software · AI infra and compute
Auralith
Over-the-counter hearing wearable with a speech model burned into custom silicon.
Auralith sells a consumer ear-worn device whose noise separation and speech reconstruction model is hardcoded into a low-power ASIC, so a full conversational model runs for a day on a coin-cell class battery with no phone and no cloud.
AI and software · AI infra and compute
Gridshift
The marketplace where AI data centers sell their flexibility to the power grid.
Gridshift turns interruptible AI training load into a tradable grid product: data centers enroll clusters, Gridshift's ML forecasts and checkpoint-aware orchestration shift or shed training jobs on minutes of notice, and the flexibility is sold into wholesale electricity markets as demand response.
AI and software · AI infra and compute
Backline
Voice and back-office agents for small businesses, running on our own inference fleet.
Backline answers the phone, books jobs, chases invoices and updates the booking system for plumbers, clinics, garages and salons, billed per handled call with no software for the owner to configure.
AI and software · AI infra and compute
Innervox
A neural wristband that turns silent speech into text for AI assistants.
Innervox builds a consumer wearable that reads surface EMG signals from subvocalized speech, letting people talk to AI assistants silently at conversational speed - in meetings, open offices, or public spaces.
AI and software · AI infra and compute
Ledgerwatt
Metering and settlement API that turns AI spend into an auditable clearing layer.
Ledgerwatt sits between enterprises and every model provider, neocloud and GPU broker they use, meters each call at the token and job level, prices it against the contracted rate card, and settles a single invoice with per-team, per-product attribution that finance can audit.
Fictional company written 2026-08-26 from MarkosWeb data; the companies, grants and numbers around it are real and tracked. Treat the idea as a research prompt, not a plan.