Alhena.ai
Hallucination-Free AI agents to turn CX into Revenue
Website: https://alhena.ai
Cover Block
PUBLIC
| Field | Value |
|---|---|
| Name | Alhena.ai (formerly Gleen AI) |
| Tagline | Hallucination-Free AI agents to turn CX into Revenue |
| Headquarters | Pleasanton, California |
| Founded | 2022 |
| Stage | Seed |
| Business Model | SaaS |
| Industry | E-commerce / Retail |
| Technology | AI / Machine Learning |
| Geography | North America |
| Growth Profile | Venture Scale |
| Founding Team | Co-Founders (2): Ashu Dubey, Surya Krishnan |
| Funding Label | Seed |
| Total Disclosed | ~$4.9M (estimated, per secondary databases) |
Links
PUBLIC
- Website: https://alhena.ai
- LinkedIn: https://www.linkedin.com/company/alhenaai
- Founder LinkedIn (Ashu Dubey): https://www.linkedin.com/in/ashudubey
- Co-founder LinkedIn (Surya Krishnan): https://in.linkedin.com/in/surya-krishnan-085aa814a
Executive Summary
PUBLIC
Alhena.ai, the Pleasanton-based company formerly known as Gleen AI, sells generative AI agents that sit on top of e-commerce stacks to handle pre-sale shopping guidance, post-sale customer support, and voice interactions, with a marketing promise of "hallucination-free" responses [Alhena.ai]. The company was founded in 2022 by Ashu Dubey and Surya Krishnan, and rebranded from Gleen to Alhena to signal a sharper focus on revenue-generating commerce use cases rather than generic support deflection [Alhena.ai]. Its differentiation is positioned at the integration layer: connecting to ERP, CPQ, CMS, PIM, and CRM systems so the agent can quote the right price, check stock, and resolve configuration questions in verticals where product complexity drives cart abandonment [Alhena.ai]. Customer logos disclosed on the company site include Puffy, SKF, DIMO, LeanData, Matter Labs (zkSync), and Tatcha, with a Puffy case study claiming 63% of inquiries resolved by the agent [Alhena.ai]. Funding to date has been seed-stage, with South Park Commons, The Spartan Group, Slow Ventures, and 6th Man Ventures named among backers [Crunchbase, 2026]. A third-party database lists revenue at approximately $4M against a 36-person team, though that figure carries lower confidence than the company-disclosed product and customer claims [GetLatka]. Over the next 12 to 18 months, the questions investors should track are whether the AI guided-selling pitch produces measurable conversion lift in named customer deployments, whether a Series A round materializes to fund go-to-market, and how the product holds up against better-capitalized support-automation incumbents.
Data Accuracy: GREEN -- Company website, Crunchbase, and LinkedIn all corroborate the core profile.
Taxonomy Snapshot
| Axis | Value |
|---|---|
| Stage | Seed |
| Business Model | SaaS |
| Industry / Vertical | E-commerce / Retail, with SaaS support adjacency |
| Technology Type | Generative AI / agentic systems |
| Geography | North America (HQ Pleasanton, CA) |
| Growth Profile | Venture Scale |
| Founding Team | Two co-founders, ex-LinkedIn / Microsoft pedigree |
| Funding | ~$4.9M disclosed across seed financings (estimated) |
Company Overview
PUBLIC
Alhena.ai was founded in 2022 under the name Gleen AI and is headquartered in Pleasanton, California [Alhena.ai]. The original framing was a generative AI platform aimed at customer support automation, with early traction concentrated in SaaS and developer-tool customers, including a public deployment with Matter Labs supporting the zkSync developer community [Alhena.ai]. CEO Ashu Dubey has described the early thesis in interviews as building a generative system whose answers could be trusted in production environments where a wrong answer carries reputational or transactional cost [Authority Magazine].
The company rebranded from Gleen to Alhena, marking a strategic pivot toward e-commerce as the primary go-to-market wedge [Alhena.ai]. The rebrand blog frames the shift as moving from generic support deflection toward AI that "actively drives success," which in practice means guided selling and conversion lift in addition to ticket resolution [Alhena.ai]. Subsequent product blogs describe vertical playbooks for B2B commerce, eyewear and optical, and configurable industrial products, suggesting the team is using vertical content as a wedge into complex catalogs where one-price DTC tools struggle [Alhena.ai].
Key milestones publicly visible include the 2022 founding, the Gleen-to-Alhena rebrand, the publication of customer case studies with Puffy and zkSync, and seed-stage financings disclosed in third-party databases [Alhena.ai] [Crunchbase, 2026]. Legal entity details and exact incorporation filings are not publicly disclosed in the captured sources.
Data Accuracy: GREEN -- Confirmed by Alhena.ai company website, Crunchbase, and founder LinkedIn profiles.
Product and Technology
MIXED
The Alhena product line, as described on the company website, comprises three agent types: AI shopping assistants for pre-purchase guidance, customer support chatbots for post-purchase service, and voice agents [PUBLIC] [Alhena.ai]. The marketing positioning centers on the phrase "hallucination-free," which the company elaborates as a combination of grounding the model in customer-specific knowledge and constraining responses to verified sources [PUBLIC] [Alhena.ai]. A third-party data page describes the system as deployable in under two hours and integrated with leading help desk solutions, though that claim originates downstream of the company's own materials [PUBLIC] [Prospeo.io].
The more interesting technical claim, and the one most relevant to e-commerce buyers, is the "AI guided selling layer" pitch: Alhena describes its agent as connecting to enterprise ERP for inventory and pricing, CPQ for configuration rules, and CMS, PIM, or CRM for content and customer context [PUBLIC] [Alhena.ai]. Vertical blog content extends this into B2B-specific use cases such as contract pricing, volume tiers, and negotiated exceptions, where the agent must pull account-specific prices from systems like SAP, Oracle, or NetSuite at query time [PUBLIC] [Alhena.ai]. This integration surface, rather than the underlying model, is where the company appears to be building defensibility.
Deployment evidence comes from two named case studies. Puffy, a US online mattress brand, is reported to have resolved 63% of customer inquiries through the Alhena agent [PUBLIC] [Alhena.ai]. Matter Labs deployed an Alhena-powered generative AI assistant to help developers working with the zkSync stack, an unusually technical customer-service surface [PUBLIC] [Alhena.ai]. The open Founding Full Stack engineering role suggests continued investment in the ticketing and integration surface, with the job description referencing "advanced ticketing solution" work and adversarial robustness testing [PUBLIC] [Alhena.ai]. The underlying model layer (whether Alhena uses third-party foundation models, fine-unes, or a proprietary stack) is not specified in the captured public sources.
Data Accuracy: YELLOW -- Product features are confirmed by company sources; deployment and integration depth rely largely on company-published case studies without independent third-party verification.
Market Research and Opportunity
PUBLIC
Generative AI agents for commerce have moved from pilot curiosity to a recognized line item in retail technology budgets, and the segment Alhena addresses sits at the intersection of two well-funded categories: customer support automation and conversion optimization.
No company-specific TAM figure was surfaced in the captured research, and Alhena does not publish its own market sizing in the cited materials. As an analogous reference, the broader AI customer support category has attracted substantial venture funding into named competitors such as Ada and Forethought, indicating institutional investor belief in the segment's scale, even where headline TAM numbers are not standardized across sources. Demand drivers Alhena's own content emphasizes are concrete and consistent with what retail operators report: complex product catalogs that drive cart abandonment, returns costs that compound when shoppers buy the wrong configuration, and the cost of human support staffing in categories with high pre-sale question volume [Alhena.ai]. The company's vertical playbooks for eyewear, B2B configurable products, and mattresses each map to categories where the question "will this work for me" is the conversion bottleneck [Alhena.ai].
Adjacent and substitute markets matter for how the opportunity is bounded. On one side sit horizontal customer support platforms (Zendesk, Intercom, Salesforce Service Cloud) that are adding generative features natively, which can either commoditize Alhena's support use case or create distribution partners. On the other side sit conversion-rate-optimization and merchandising tools that are also racing toward agentic experiences. Alhena's chosen position, an agent that talks to ERP, CPQ, PIM, and CRM at the same time, sits between those two camps and is, on paper, harder to replicate from either starting point [Alhena.ai]. Regulatory and macro forces include consumer protection rules around AI-generated claims, where the "hallucination-free" framing functions as both marketing and risk mitigation, and the broader push by retailers to reduce return rates as logistics costs remain elevated.
| Sizing reference | Value | Source |
|---|---|---|
| Alhena disclosed revenue (single-source) | ~$4M | [GetLatka] |
| Alhena headcount | 36 | [GetLatka] |
| Alhena disclosed funding | ~$4.9M (estimated) | [Crunchbase, 2026] |
The table summarizes the only quantified data points captured for Alhena itself; all three originate in third-party databases rather than primary disclosures, and the revenue figure in particular should be treated as directional rather than audited.
Data Accuracy: ORANGE -- Market context is reasoned from analogous segments; company-specific quantitative figures rely on a single secondary database.
Competitive Landscape
MIXED
Alhena competes in a crowded AI customer experience category but has chosen a narrower position inside it: agentic guided selling for complex e-commerce catalogs, rather than horizontal support deflection.
| Company | Positioning | Stage / Funding | Notable Differentiator | Source |
|---|---|---|---|---|
| Alhena.ai | Hallucination-free AI agents for e-commerce shopping, support, and voice | Seed, ~$4.9M disclosed (estimated) | Integration depth into ERP / CPQ / PIM / CRM for guided selling | [PUBLIC] [Alhena.ai] [Crunchbase, 2026] |
| Ada | Conversational AI customer service automation | Late-stage, multi-hundred-million raised per public reporting | Enterprise support breadth across industries | [PUBLIC] |
| Forethought | Generative AI for customer support agents and ticket triage | Series C, multi-round venture-backed per public reporting | Agent-assist and triage focus inside support orgs | [PUBLIC] |
The segment-by-segment map looks like this. Incumbent support-automation platforms (Ada, Forethought, plus the embedded AI features inside Zendesk and Intercom) own the deflection budget at large enterprises. Horizontal foundation-model players and their orchestration partners (custom GPT builds, LangChain-based implementations from agencies) compete at the low end on price and flexibility. Adjacent substitutes include merchandising and personalization tools moving toward conversational interfaces, and the in-house build option that any sufficiently large retailer can pursue. Alhena's chosen wedge, agents that resolve questions requiring real-time pricing, inventory, and configuration data, is a thinner slice and less directly contested by the deflection-first incumbents [Alhena.ai].
Where Alhena appears to have a defensible edge today is in the integration surface and the vertical content motion. Connecting cleanly to ERP and CPQ systems is a non-trivial implementation lift that, once deployed, creates switching costs because the customer's product and pricing logic ends up reflected in the agent's grounding [Alhena.ai]. The vertical blogs for B2B, eyewear, and configurable industrial products suggest a content-led inbound motion that targets buyers whose pain (cart abandonment driven by complexity) is concrete enough to fund a paid pilot. Whether that edge is durable depends on how quickly Ada, Forethought, and Salesforce close the same integration gaps, which is a question of capital and partnerships rather than fundamental technology.
Where Alhena is most exposed is enterprise distribution. Ada and Forethought have multi-year head starts in the procurement workflows of large support organizations, and Salesforce can bundle Service Cloud AI into existing contracts. A second exposure is the model layer: if commodity foundation models continue improving on grounding and retrieval, the "hallucination-free" framing becomes less differentiated and the value migrates either upward to workflow or downward to data integration. The most plausible 18-month scenario splits two ways. Winner if Alhena converts its named customers (Puffy, Tatcha, SKF, LeanData, DIMO, zkSync) into multi-product, multi-year reference accounts that anchor a Series A and an enterprise sales build-out [Alhena.ai]. Loser if the company stays sub-$10M ARR while a better-capitalized competitor ships a credible guided-selling SKU and absorbs the integration playbook through partnership or acquisition.
Opportunity
PUBLIC
If Alhena executes, the prize is to become the default agentic layer between complex product catalogs and the shoppers trying to navigate them, a position that sits one step above support automation and one step closer to revenue.
The headline opportunity. The most ambitious plausible outcome for Alhena is to become the standard guided-selling agent for mid-market and enterprise e-commerce operators in categories where product complexity is the primary conversion barrier. The cited evidence makes that outcome reachable rather than aspirational on three counts: a working product with named reference customers across mattresses, industrial bearings, beauty, and developer tools [Alhena.ai]; an integration story (ERP, CPQ, PIM, CRM) that maps to where retail technology budgets actually flow [Alhena.ai]; and a founding team whose backgrounds include LinkedIn, Microsoft, Facebook, Amazon, and McKinsey, which is the kind of pedigree that opens enterprise doors [Alhena.ai] [Prospeo.io]. The category-defining version of Alhena is the company that retailers reach for when they accept that catalog complexity, not chatbot deflection, is the problem worth solving.
Growth scenarios.
| Scenario | What happens | Catalyst | Why it's plausible |
|---|---|---|---|
| Vertical land-and-expand | Alhena dominates two or three complex-catalog verticals (mattresses, eyewear, configurable B2B industrial) and becomes the reference agent for each | A flagship case study showing measurable conversion lift, not just inquiry resolution, in a named brand | Existing Puffy and SKF deployments already span DTC and industrial complexity [Alhena.ai] |
| Embedded commerce layer | Alhena becomes a packaged add-on inside one or more commerce platforms (Shopify Plus apps, BigCommerce, Adobe Commerce, SAP Commerce Cloud) | A platform partnership or marketplace listing that makes Alhena the recommended guided-selling agent | The integration surface (ERP, CPQ, PIM) is exactly what platform partner programs reward [Alhena.ai] |
| Voice and agentic commerce wedge | The voice agent line becomes the entry product for retailers preparing for agentic checkout standards | A consumer-side agentic commerce protocol gaining traction, requiring merchants to expose structured product and pricing data | Alhena already ships a voice agent and publishes content on agentic commerce readiness [Alhena.ai] |
What compounding looks like. The flywheel that makes Alhena interesting is integration data. Each ERP, CPQ, or PIM connector that the team builds for one customer reduces the deployment cost for the next customer in that vertical, which compresses sales cycles and improves gross margin. The vertical content library (B2B pricing, eyewear, mattress, agentic commerce readiness) is a parallel inbound flywheel that lowers customer acquisition cost as the corpus grows [Alhena.ai]. Early signs the flywheel is starting include the published 48-hour deployment narrative and the breadth of customer logos already spanning DTC, industrial, beauty, and crypto-developer categories [Alhena.ai]. Where compounding could fail to materialize is if each enterprise deployment remains bespoke enough that connector reuse never crosses the threshold where gross margin expands.
The size of the win. Public peers in the AI customer experience category have raised at valuations well into nine and ten figures, and the broader e-commerce platform category contains multiple multi-billion-dollar outcomes. A scenario where Alhena reaches $50M to $100M ARR as the default guided-selling agent in two or three complex verticals would, on category-comparable revenue multiples, support a venture-scale outcome consistent with what its current backers underwrite (scenario, not a forecast). The more interesting upside case is the platform-embedded outcome, where Alhena becomes a strategic acquisition target for a commerce platform or support incumbent that needs a guided-selling story; that pathway typically clears at a premium to standalone revenue multiples. Both outcomes depend on the next 18 months of customer expansion and a credible Series A that funds enterprise distribution.
Data Accuracy: YELLOW -- Scenario logic is grounded in named customers and product claims from company sources; valuation comparables are illustrative rather than independently sourced for Alhena specifically.
Sources
PUBLIC
[Alhena.ai] Alhena AI | Hallucination-Free AI agents to turn CX into Revenue | https://alhena.ai
[Alhena.ai] About Alhena (Formerly Gleen) | https://alhena.ai/about
[Alhena.ai] Generative AI Chatbot for SaaS BigTech | https://alhena.ai/solutions/saas-big-tech
[Alhena.ai, 2026] AI Guided Selling for Complex Product Configuration | https://alhena.ai/blog/ai-guided-selling-product-configuration/
[Alhena.ai] AI for Eyewear & Optical Ecommerce: The Complete Guide | https://alhena.ai/blog/ai-eyewear-optical-ecommerce/
[Alhena.ai] Gleen AI is Now Alhena AI: Revolutionizing eCommerce | https://alhena.ai/blog/gleen-ai-is-now-alhena-ai/
[Alhena.ai] How Puffy used Alhena AI to Resolve 63% of Customer Inquiries | https://alhena.ai/case-studies/puffy
[Alhena.ai] zkSync Introduces Generative AI Solution Powered by Alhena AI | https://alhena.ai/case-studies/zksync
[Alhena.ai] Careers at Alhena: Founding Full Stack | https://alhena.ai/jobs/founding_full_stack
[Alhena.ai] AI for B2B Ecommerce: Pricing, Bulk Orders & Accounts | https://alhena.ai/blog/ai-b2b-ecommerce-chatbot/
[Alhena.ai] What Happens in the First 48 Hours: Alhena AI Setup | https://alhena.ai/blog/alhena-ai-48-hour-setup-ecommerce/
[LinkedIn] Ashu Dubey - CEO @ Alhena.ai | https://www.linkedin.com/in/ashudubey
[LinkedIn] Surya Krishnan - Alhena.ai (formerly Gleen AI) | https://in.linkedin.com/in/surya-krishnan-085aa814a
[LinkedIn] Alhena.ai company page | https://www.linkedin.com/company/alhenaai
[GetLatka] How Alhena.ai hit $4M revenue with a 36 person team | https://getlatka.com/companies/alhena.ai
[Prospeo.io] Alhena.ai Revenue, Funding & Valuation | https://prospeo.io/c/alhena-ai-revenue
[Tracxn] alhena.ai 2025 Company Profile | https://tracxn.com/d/companies/alhena.ai/__YDu4KUePiKoAOgcJrqd_5NQeFTHdTfXYJMU0aGX6lwc
[Authority Magazine on Medium] Ashu Dubey of Gleen On Five Things You Need To Create A Highly Successful Career In The AI Industry | https://medium.com/authority-magazine/ashu-dubey-of-gleen-on-five-things-you-need-to-create-a-highly-successful-career-in-the-ai-industry-c5ce958b5f2d
[AI Workplace Wellness Substack] Retaining Marketing's Humanity in an AI-Powered World: Interview with Ashu Dubey | https://aiworkplacewellness.substack.com/p/retaining-marketing-humanity-in-an
Articles about Alhena.ai
- Alhena.ai Is Selling Hallucination-Free Shopping Agents to Mattress Brands and Crypto Devs — The Pleasanton startup, formerly Gleen, says its AI resolved 63% of Puffy's customer inquiries. Now it wants the complex-product corner of e-commerce.