Is BaseFrame profitable in 2026?

No public revenue, ARR, or profitability figures exist for BaseFrame as of May 2026. The company is a two person, pre revenue Y Combinator W26 startup; an unattributed claim of a W26 company hitting $27M ARR does not name BaseFrame specifically.

What does BaseFrame actually do?

As of May 30, 2026, BaseFrame is a desktop app that scans email, docs, Slack, calendar, and CRM history using local small AI models to surface 5 to 10 automatable tasks per employee, then hands them off to whichever automation tool the customer already uses.

Could AI replace BaseFrame?

Yes, plausibly. BaseFrame's core function, detecting automatable tasks from usage data, is exactly what Microsoft Copilot, Google Workspace AI, and Salesforce Agentforce could do natively since those platforms already own the same email, calendar, and CRM data as of mid 2026.

Who are BaseFrame's biggest competitors?

As of 2026, BaseFrame competes with venture backed task mining vendors Skan AI, Mimica, Soroco, and KYP.ai, plus process mining leader Celonis and newer agent native rivals Worktrace and Fluency, all pursuing a similar automation discovery thesis.

Has BaseFrame changed its product before?

Yes. BaseFrame launched a hardware design risk tool around January 16, 2026, pivoted to an AI agent security product called Clam by March 8, 2026, then relaunched as an automation discovery app by May 30, 2026, two full pivots in about four months.

BaseFrame is privately held and venture backed through Y Combinator's Winter 2026 batch, founded by CEO Anshul Paul and co-founder CTO Vaibhav Agrawal in 2025. No named seed round size, valuation, or outside lead investor has been publicly disclosed as of mid 2026.

YC Winter 2026 · United States

BaseFrame

App that scans email, Slack, and docs to flag tasks AI could automate.

Little protection left

Assessed 27 July 2026 · Evidence as of 27 July 2026

Not much stands between this business and its pressures.

On the recordSince 27 July 2026 · Checkable from H2 2026 · Every call on the record Checkpoint: Microsoft and Google expected to expand native Copilot/Workspace AI task-detection features

The call

A two-person team scanning email and Slack for automatable tasks, sitting on data and rails that Microsoft, Google, and RPA vendors already control.

BaseFrame shows companies which repetitive tasks in their daily work could be automated. The problem: Microsoft, Google, and Salesforce are adding this same feature directly into their own products that companies already use, and tools like ChatGPT with email and document access already do this analysis. BaseFrame also owns no data and builds nothing itself, just points to other tools. The company changed what it was building three times in four months, which suggests the founders are still searching for a real problem to solve, and that lack of focus is the biggest warning sign.

Why

AI pressure: Microsoft Copilot and Claude/ChatGPT connectors into email, Slack, and Drive. Copilot and Claude already read the same inboxes and files to suggest automations, matching BaseFrame's core pitch for free. Could AI do the core job? Yes, likely.

Bigger players: UiPath Task Mining, Automation Anywhere Discovery Bot, Salesforce Agentforce. RPA vendors already ship task-mining inside platforms companies already pay for, leaving little reason to add a separate scanner. Could a giant copy it? Yes, easily.

What protects it. A SOC 2 Type I badge and a zero-retention privacy pitch, aimed at winning trust from early adopters before bigger players fold this in.

What would change this call

Microsoft or Google adds native task-discovery to Copilot or Workspace AI in H2 2026. Would eliminate BaseFrame's discovery-layer differentiation using data it does not own.
BaseFrame announces a third product pivot before Q4 2026. Would confirm a pattern of repositioning and lack of conviction on this problem.
Celonis, Soroco, or a major RPA vendor acquires a task-mining startup within 12 months. Would show consolidation compressing BaseFrame's competitive window.

Key numbers

Two product pivots in four months, Jan to May 2026 (2026-05-30)
Two-person founding team (n.d.)
10 integrations listed, including Gmail, Slack, Salesforce (2026-05-30)

Named risks

Two full product pivots in about four months signal weak founder conviction on this specific problem. Severity: high
No proprietary data or execution ownership; hands automations off to third party tools, leaving no lock-in. Severity: high
No verified revenue, ARR, or customer count as of May 2026 despite public product launch. Severity: medium
Category already crowded with 5+ funded specialists and two major RPA incumbents before product stabilized. Severity: medium
Platform giants sit natively on the same usage data BaseFrame needs, with no data or regulatory moat blocking absorption. Severity: high

Rivals, as read in July 2026

Rival	Direction	Evidence
UiPath and Automation Anywhere	gaining on	RPA incumbents already ship task mining/discovery modules, sitting closer to enterprise deployment than BaseFrame
Skan, Mimica, Soroco, KYP.ai	contesting	5+ venture backed task mining vendors capture desktop activity for automation candidates as of 2026-06-15
Celonis	losing to	Market leading incumbent in enterprise process mining, the reference point rivals position against as of 2025-05-23
Microsoft, Google, Salesforce	gaining on	Copilot, Workspace AI, and Agentforce natively own the same email, docs, and CRM data BaseFrame accesses via integrations
Worktrace and Fluency	contesting	Newer agent native workflow discovery startups pursuing the same thesis as of 2026-02-04

Questions people ask

Is BaseFrame profitable in 2026?: No public revenue, ARR, or profitability figures exist for BaseFrame as of May 2026. The company is a two person, pre revenue Y Combinator W26 startup; an unattributed claim of a W26 company hitting $27M ARR does not name BaseFrame specifically.
What does BaseFrame actually do?: As of May 30, 2026, BaseFrame is a desktop app that scans email, docs, Slack, calendar, and CRM history using local small AI models to surface 5 to 10 automatable tasks per employee, then hands them off to whichever automation tool the customer already uses.
Could AI replace BaseFrame?: Yes, plausibly. BaseFrame's core function, detecting automatable tasks from usage data, is exactly what Microsoft Copilot, Google Workspace AI, and Salesforce Agentforce could do natively since those platforms already own the same email, calendar, and CRM data as of mid 2026.
Who are BaseFrame's biggest competitors?: As of 2026, BaseFrame competes with venture backed task mining vendors Skan AI, Mimica, Soroco, and KYP.ai, plus process mining leader Celonis and newer agent native rivals Worktrace and Fluency, all pursuing a similar automation discovery thesis.
Has BaseFrame changed its product before?: Yes. BaseFrame launched a hardware design risk tool around January 16, 2026, pivoted to an AI agent security product called Clam by March 8, 2026, then relaunched as an automation discovery app by May 30, 2026, two full pivots in about four months.
Who owns BaseFrame?: BaseFrame is privately held and venture backed through Y Combinator's Winter 2026 batch, founded by CEO Anshul Paul and co-founder CTO Vaibhav Agrawal in 2025. No named seed round size, valuation, or outside lead investor has been publicly disclosed as of mid 2026.

Timeline

2026-05-30Live site shows current product: desktop app detecting AI automatable tasks from work history
2026-03-08Co-founder Vaibhav Agrawal discusses prior pivot to Clam, an AI agent security product
2026-02-02Coverage describes original hardware design risk detection concept
2026-01-16BaseFrame publicly launches on YC as a hardware design confidence and BOM tool
2025BaseFrame founded by Anshul Paul and Vaibhav Agrawal

Revision history

27 July 2026

First assessed. Little protection left. The text above is fixed as of this date and is not edited afterwards.

In the same band

GRU Space Agentic Fabriq Vela