Aiolum Browse stories
All stories
AI agents that implement enterprise software (ERP, CRM, ITSM) for software vendors and system integrators · India

Lab0: Product, team and how it was built

Small teamFundedB2BEnterprise software

Lab0 (AI agents that implement enterprise software (ERP, CRM, ITSM) for software vendors and system integrators, India): Live AI implementation agents from YC Spring 2026, November 2026. Evidence grade E: no revenue figure was checked. Team: 3 founders at the Y Combinator launch in May 2026; team size 7 on the Y Combinator company page, November 2026. Launched May 2026; still running.Added 6 November 2026

Outcome
Live AI implementation agents from YC Spring 2026, November 2026
Evidence
E no revenue figure was checked
Launch
May 2026
Status
Still running
Team
3 founders at the Y Combinator launch in May 2026; team size 7 on the Y Combinator company page, November 2026
Outside funding
Y Combinator, Spring 2026 batch; amount not published
Main channel
Not stated
No revenue figure

What is not verified

No revenue figure for Lab0 was found. The lead came from Arrfounder, a founder directory that shows US$12K ARR on Lakshya Gupta's card (collected June 2025); its product pages tie that figure (US$1K MRR, extracted August 2025, source 'ai') to Cvbrew, an AI resume builder listed under his handle, not to Lab0, which was founded in 2026. That figure is therefore not used. No payment-processor profile, filing or interview with Lab0 numbers was found. The Adobe relationship rests only on the launch post. A reader could look for a seed-round announcement or a customer case that names a client. This story is grade E: the business is real and its product, price and customers are described from public pages, but no revenue figure was checked. Aiolum shows it as an idea worth studying, not as proof that it makes money.

Lab0 builds what it calls an AI FDE, an AI forward-deployed engineer: software that does the work of the engineer a vendor sends to a customer to turn bought software into a working deployment. That work is implementation, the setup of a product for one company: discovery of its processes, configuration of fields, forms and rules, integration with existing systems such as an ERP (enterprise resource planning, the finance and operations system of record, for example SAP), and testing before go-live.

Onkar Borade (CEO), Lakshya Gupta (CTO) and Sujay Srivastava founded it in 2026 and launched it on Y Combinator on 20 May 2026 as part of the Spring 2026 batch. The Y Combinator page lists Bengaluru, India, and a team size of 7. The launch post says the company was working with Adobe and "a couple of system integrators"; no customer confirmed this.

No revenue figure for Lab0 was found, so this story is grade E: the product is live and sold through discovery calls, but its money is not verified. The page also shows earlier companies that tried to remove custom integration work between business systems and failed.

Lab0: Live AI implementation agents from YC Spring 2026, November 2026. AI agents that implement enterprise software (ERP, CRM, ITSM) for software vendors and system integrators, India, team of 2 to 10, evidence grade E.
The story in one card. Share it with the address of this page.
01

How much revenue does Lab0 make?

Each row shows the evidence grade and the source numbers from the list at the end of this page. The grades are explained in How Aiolum verifies.

A seven-person Y Combinator company selling AI agents for enterprise software implementation, with no published revenue
FactValueGradeHow we knowSources
Status, November 2026Live site, sales through discovery callsCHome page read 6 November 2026: 'AI accelerators for ERP implementations', 'Book a discovery call', 'Backed by Combinator' (sic).[1]
FoundersOnkar Borade (Founder/CEO), Lakshya Gupta (Founder/CTO), Sujay Srivastava (Founder)CY Combinator company page founder list and launch post 'Who we are': all three studied at IIT Bombay or IIT Madras; Borade worked at Oracle.[4, 5]
Team7 people, November 2026CY Combinator company page data: team_size 7. Started by 3 founders (launch post), with outside funding from Y Combinator, so the 6 to 10 rule for funded teams applies.[4, 5]
Founded and location2026, Bengaluru, IndiaCY Combinator company page: year_founded 2026, location 'Bengaluru, India', batch 'P26'.[4]
Launch20 May 2026, Y Combinator launch postCLaunch 'Lab0 - AI FDE for Enterprise Software', created 2026-05-20, 20 votes when read.[5]
FundingY Combinator, Spring 2026 batch; amount not publishedCY Combinator page lists batch P26; the site says 'Backed by Y Combinator'. No source gives an amount or a later round.[2, 4]
AgentsDiscovery, Artifact, Planning, Configuration, Testing, Change RequestCSite menu. Changes run as a 'controlled dry-run' before they write to a live customer instance; approval and rollback stay with the customer's team (AI FDE page).[1, 2]
ConnectorsSAP, Oracle ERP, Workday HCM, Microsoft Dynamics, Salesforce, ServiceNowCSite menu 'Connectors'.[1]
Claimed speed-upabout 26 weeks and 9 people to about 4 weeks and 3 peopleCCompany's own worked example on the 'How Lab0 works' page (a fictional company 'Acme' implementing Coupa with ServiceNow and SAP). Not a measured customer result.[3]
Example deploymentUAE e-invoicing across ERPs (Business Central, Oracle Fusion, SAP Business One, NetSuite, QuickBooks)CCase page: a 'native extension' for Business Central for e-invoicing providers. The page names no customer.[6]
Time from launch to the November 2026 status6 monthscalcCalculated: May 2026 to November 2026.[5]

How to read this table: product values come from the Lab0 site read on 6 November 2026; founders, launch date and team size from the Y Combinator company page and launch post read the same day; the directory figure from Arrfounder pages that the pipeline downloaded on 6 November 2026. No value is grade A or B.

Lab0 prices in November 2026: none published (grade C)
PlanPer month, billed yearly (US$)Per month, billed monthly (US$)Limit
Contract after a discovery callno prices published

The site has no pricing page. Every page leads to "Book a discovery call" (a Calendly link), and the launch post offers to "take one of your hardest customer deployments and run it end-to-end".

02

When did Lab0 launch, and how did it grow?

From a Y Combinator launch post in May 2026 to a live site with six agents and an e-invoicing case in November 2026. Each entry carries its grade.

  1. 2026COnkar Borade, Lakshya Gupta and Sujay Srivastava found Lab0 in Bengaluru. [4]
  2. 20 May 2026CY Combinator launch post: 'Deploy enterprise software in days, not months'; it says Lab0 works with Adobe and a couple of system integrators. [5]
  3. May 2026CLab0 is listed in the Y Combinator Spring 2026 batch. [4]
  4. November 2026CThe site is live with six agents, six platform connectors, a ServiceNow demo and a UAE e-invoicing case; the Y Combinator page lists 7 people. [1, 4, 6]
03

Why did Lab0 work?

Four causes, in the order they mattered. Each one is a reusable lesson.

01

Sell to the vendor, not the end customer

Lab0 sells to software companies and system integrators that already carry implementation work, so one contract covers many end-customer rollouts. The home page names both buyer types.

02

Turn repeated services work into a product

The AI FDE page argues that human forward-deployed engineers 'don't scale' and that each rollout repeats the same discovery, configuration and test work; Lab0 runs one standard agent workflow and keeps what it learns for the next rollout.

03

Safety by dry-run

Every change is first applied as a preview and checked against test cases; the customer's team keeps approval and rollback. Enterprise buyers need this before an agent may write to a live ERP.

How it was built

The sources describe the product design, not the technical stack. [2, 6, 1]

  • Six agents for the phases of an implementation: discovery, artifacts (design documents), planning, configuration, testing and change requests.
  • Connectors to SAP, Oracle ERP, Workday, Microsoft Dynamics, Salesforce and ServiceNow.
  • A 'deterministic-versus-AI split' for the places where human judgment is needed, and dry-run previews before any write.
  • A native Business Central extension for the UAE e-invoicing case.
  • The marketing site is built with Astro; discovery calls are booked through Calendly.

Tools named in the sources

Each one opens its page in the Tools tab, with every story that uses it.

04

Who tried this before Lab0, and why did they fail?

Two companies in our database of failed startups tried to remove custom integration work between business systems and failed. The values below are from that database and are not independently verified.

Died 2017 to 2025Burned US$81M

Solid

The promise

One unified API to connect CRMs, ERPs, HR systems and other enterprise tools without custom integration work.

Why it died

It underestimated how complex enterprise integration is; high customer acquisition cost, long sales cycles and support costs broke its unit economics.

Unit Economics
Died 2018 to 2021Burned US$2M

Xkit

The promise

A platform that made connecting and managing third-party API integrations simpler for product teams.

Why it died

Zapier and do-it-yourself cloud tools left it little room, and it could not keep up with its own burn rate.

Competition

The failed startups come from our database of 1,749 dead startups, retold here in our own words. Its entries are AI-assisted summaries of public sources, so these cards carry no evidence grade.

What changed

Solid and Xkit tried to make integration a fixed API layer, and Solid found that each enterprise needs custom work that a fixed layer cannot cover. Lab0 does not avoid the custom work; it has agents do it per customer, which language models made possible because they can read requirement documents and write configuration and test cases.

The risk that killed Solid, long enterprise sales cycles, still applies: Lab0 sells through discovery calls to large software vendors.

05

Can one person copy Lab0 now?

Hard for one person; it needs enterprise sales and deep knowledge of each platform.

The method is to find a services step that every enterprise sale repeats, encode the steps as agents with safe previews, and sell to the vendors who carry that cost. Lab0's founders came from inside that work (Oracle implementation, enterprise AI rollouts), which is most of the moat.

Where it is still open: Other repeated post-sale work: data migration between CRMs, security questionnaire answers for each enterprise deal, and localisation of tax rules in ERPs.

  1. 1Pick one platform and one repeated project type (for example e-invoicing setup in one ERP).
  2. 2Write agents for discovery, configuration and testing, and make every write a reviewable dry-run.
  3. 3Sell to the software vendor or integrator that does this work today, and run its hardest deployment as proof.
  4. 4Add connectors one platform at a time.

What could go wrong for Lab0?

The risks are the mirror image of the causes above.

What is not verified

No revenue figure for Lab0 was found. The lead came from Arrfounder, a founder directory that shows US$12K ARR on Lakshya Gupta's card (collected June 2025); its product pages tie that figure (US$1K MRR, extracted August 2025, source 'ai') to Cvbrew, an AI resume builder listed under his handle, not to Lab0, which was founded in 2026. That figure is therefore not used. No payment-processor profile, filing or interview with Lab0 numbers was found. The Adobe relationship rests only on the launch post. A reader could look for a seed-round announcement or a customer case that names a client.

Services or software

The site says the AI FDE is delivered 'as a product, not a person', but sales start with a discovery call and the launch post offers to 'take one of your hardest customer deployments and run it end-to-end'. How much of each rollout is done by Lab0 staff is not stated.

Team size and location

The Y Combinator page lists 7 people and Bengaluru, India; the CTO's directory card lists San Francisco. The headcount is above the site's small-team range of 2 to 5 and is allowed under the rule for funded teams started by 1 to 5 people.

Competition

Lab0 competes with system integrators and with agents built into each platform (its AI FDE page compares itself with in-platform agents such as Palantir Foundry's); on this site the closest stories are developer-tools stories about agents for engineers, none of which sell enterprise implementation.

Dependence on model providers

No downloaded source names the language models Lab0 uses or a model provider, so its dependence on one provider cannot be judged.

Sources

The grade next to each source is the best grade that the source can support. "Read in full" means that the text of the page was downloaded and the numbers were checked in it.

  1. C lab0: AI accelerators for ERP implementations (home page)
    Read in full on 6 November 2026. Source for the agents, connectors, buyer types, the Y Combinator badge and the Calendly booking link.
  2. C What is an AI FDE? | lab0
    Read in full on 6 November 2026. Source for the AI FDE definition, the dry-run safety model and the comparison with in-platform agents.
  3. C How Lab0 works, AI forward-deployed engineer
    Read in full on 6 November 2026. Source for the worked example (about 26 weeks and 9 people versus about 4 weeks and 3 people).
  4. C Lab0: AI FDEs to make your B2B SaaS Go Live in Days, not Months | Y Combinator
    Read the page data on 6 November 2026. Source for founders and titles, team size 7, year founded 2026, Bengaluru and batch P26.
  5. C Lab0 - AI FDE for Enterprise Software (Y Combinator launch, 20 May 2026)
    Read in full on 6 November 2026. Source for the launch date, the founders' backgrounds and the Adobe claim.
  6. C UAE E-Invoicing Implementation Across ERPs | Lab0
    Read in full on 6 November 2026. Source for the e-invoicing case and the supported ERPs.
  7. D Arrfounder: $10K-100K ARR Founders, page 3
    Downloaded by the pipeline on 6 November 2026; read the card for @heat_bender (Lakshya). Directory data; source for the US$12K ARR card figure that is not used.
  8. D Arrfounder: $10K-100K ARR Products, page 4
    Downloaded by the pipeline on 6 November 2026; read the record that ties the US$12K ARR figure to Cvbrew under the same handle.
  9. C Database of failed startups, local export
    Local export. AI-written summaries, not independently verified. Source for the failed predecessors.

Run this company? Send corrections or a verified figure to aiolum@livyden.com

Back to all stories