Firmly Lands JPMorgan and Mastercard to Rewire Retail for AI Agents

Merchants need a new on-ramp

Firmly, a Seattle-area startup founded in 2019, closed a $15 million Series A led by Squarepoint Capital, with J.P. Morgan Payments, Mastercard, and Samsung Next among the participants, as GeekWire reported ^[1]. The company's core thesis is that retailers shouldn't have to build a separate integration every time a new AI shopping channel appears — Firmly acts as the connective layer between existing retail infrastructure and the growing universe of AI assistants through which consumers are beginning to browse and buy.

The strategic weight of this round is worth noting. J.P. Morgan Payments and Mastercard are not typical venture investors chasing narrative; they participate in rounds where the commercial plumbing is directly relevant to their networks. Their presence here suggests that major payment infrastructure players view AI shopping agents not as a distant trend but as an imminent channel requiring standardized connectivity — and that they are willing to back the companies building that standard early.

For founders raising right now, this is a template worth studying. The round is not enormous in dollar terms, but its composition — a quant-rooted lead fund plus two global financial institutions — reflects a clear institutional logic. When your product sits at the intersection of existing transaction infrastructure and an emerging interaction layer, strategic capital from incumbents can signal credibility more loudly than the round size alone.

The scale gap at the top of the AI market

Zoom out from Firmly's $15 million and the contrast with the top of the market is stark. Fal, an AI video, image, and audio model startup based in San Francisco, raised $1.3 billion at a $16 billion valuation, led by Bessemer Venture Partners with Nvidia, Sound Ventures, and T. Rowe Price joining, as Axios reported ^[2]. At the same time, Nooks, a revenue agent platform, pulled in $500 million from General Catalyst's Customer Value Fund, and Applied Compute raised $350 million at a $3.25 billion valuation for custom open-source enterprise AI models ^[2].

These are not seed experiments. They are institutional-scale bets on AI infrastructure — the model layer, the agent layer, the compute layer. The pattern that has been building across recent weeks is now difficult to ignore: capital is concentrating at two ends of the market. Foundational AI infrastructure companies are absorbing rounds in the hundreds of millions, while focused application-layer startups like Firmly are raising more modest rounds but attracting strategically significant backers. The middle — the unfocused Series A with a generic AI wrapper — is where the funding environment is least forgiving.

Fintech continues to clear the bar

Fintech remained active this week. Pliant, a Berlin-based B2B payment solutions company, raised $80 million in a Series C led by Seaya Nova and Speedinvest, with Illuminate Financial, Motive Partners, and Molten Ventures participating, per Axios ^[5]. Kalepa, which applies AI to commercial insurance underwriting, raised $75 million in a Series B at a $470 million valuation, led by Updata Partners ^[5].

Kalepa's round is particularly instructive for founders working at the intersection of AI and regulated industries. Insurance underwriting is a domain with decades of proprietary data, complex risk modeling, and high switching costs — exactly the kind of vertical where AI applied with genuine domain depth commands a premium valuation. A $470 million valuation on a Series B, in an industry not known for fast cycles, points to how much investors are willing to pay for defensible vertical AI with a credible path to replacing legacy processes.

Global signals worth watching

Emerging market deal velocity remained strong. According to Entrackr, 28 Indian startups raised a combined $258.1 million across the week of November 2–7, spread across three growth-stage deals and 17 early-stage deals ^[3]. The breadth of early-stage activity in that number — 17 deals — matters as much as the aggregate figure. It suggests the pipeline of new company formation in the market is active, not just the later-stage follow-on cycle.

In Europe, Earlybird closed its Fund VIII at €360 million, with an explicit focus on deeptech and what the firm describes as patient capital, as reported by Mam Startup ^[4]. The fund's historical portfolio includes N26, UiPath, and Isar Aerospace. The closure of a fund of this size, oriented toward longer development timelines in hard technology, is relevant context for European founders building in areas — defense tech, industrial AI, biotech infrastructure — where the right investor posture is measured in years, not quarters. The fund closed earlier in 2026 but its composition reflects durable LP appetite for the category.