The concept and terminology of a “thin market” is fairly newly minted, but the reality of trying to operate in one is extremely real. Moreover, the connection between AI (LLM+RAG) technology and building systems to actualize thin markets is already well underway. As the examples in the following table illustrate, Deeperpoint and Cosolvent are not inventing a new concept. Lots of startups are already building new markets in very thin areas using AI tools. To be clear, the companies listed below don’t use any Cosolvent technology and they are likely unaware that Cosolvent exists. However, that’s sort of the point.
Despite the fact that these are independent initiatives (and independent from Cosolvent), there are striking similarities among these startups, both in the structure of the markets they are addressing and the selection of AI tools that they have chosen to overcome the market frictions.
| Logo | Description | Thin Market Characteristics | Cosolvent-Like Features |
|---|---|---|---|
 | Boardy is an AI-powered professional networking platform. It uses WhatsApp and voice AI to connect professionals (founders, investors, consultants) through conversational interviews using WhatsApp. | Thin market: Angel investors and startup founders want to find each other, but can’t do so easily or reliably. | Use AI to build user profiles from unstructured (audio interview) information Use AI to filter users with similar or matching interests. |
 | BReady – First AI-powered hiring platform connecting newcomers and talent from underinvested communities to meaningful employment opportunities with SMEs and startups across the province. | Thin market: Some Canadians have deep experience but they can’t display it efficiently. Employers would love their skills, but can’t spot them. | Talent uploads profile once and AI matches with employer needs. AI generates short lists for employers |
 | Torg – AI-powered platform connecting buyers with 100,000+ verified food/beverage suppliers in Europe. Uses AI assistants “Alex” (for suppliers) and “Sofia” (for buyers) to automate matching. | Thin market: European food and beverage industry is huge, diverse, multicultural and multilingual. Finding the best product or deal is a challenge. | AI understands buyers needs AI continuously scans suppliers for opportunities and looks for best deals |
 | Pietra – AI-powered platform connecting e-commerce businesses with 1,200+ vetted manufacturers globally. Also has “Retail Connect” – an AI assistant matching small businesses with local retail shops. | Thin market: Small businesses don’t have the bandwidth to search widely for the best suppliers. They often must do with old standbys and big, well known suppliers. | AI finds factories that match small business needs |
 | HealthKey – Prescreens Patients to connect the EHR to analyze patients’ data in a HIPAA compliant way, prescreen and encode trials’ eligibility criteria and scan EHR for candidates using AI, and review matches with evidence from patients’ records so the research team can qualify them. | Thin market: It’s hard for clinical researchers to find participants for their trials. Even if they find a candidate, there many things to consider before they can conclude it is a suitable match. | Use AI to extract matchmaking criteria from existing information. Use AI to search for matches that fit complex criteria. |
 | GreatBuildz – Matching homeowners with contractors + AI bid comparison BidCompareAI tool automatically parses and evaluates contractor bids, providing side-by-side comparisons | Thin market: Homeowners struggle to compare contractors and understand bid inconsistencies | AI connects homeowners with a shortlist of contractors that fit budget and needs |
 | Faktorist – AI-powered platform connecting verified manufacturers with international buyers. It is currently focused on helping Ukrainian companies. | Thin market: Ukraine is on the edge of Europe and embroiled in a war. Its companies need to export and import, but finding good matches is challenging. | AI matching based on specifications, capacity and requirements. Secure messaging with AI documentation assistance. |
 | Investor Match – Revolutionizing funding by leveraging artificial intelligence to form data based connections between funders, founders, and vendors based on hundreds of criteria. Streamline the search process to save precious time, resources, and effort. | Thin market: Angel investors and startup founders want to find each other, but can’t do so easily or reliably. | AI to match and form connections based on many criteria. AI optimizes the match making process |
 | Scinapse – AI Expert Finder analyzes publications, citations, h-index, institutional affiliations to identify potential collaborators | Thin market: Scientific knowledge has become so specific and experts so widely dispersed that finding an expert in a given subject is difficult. | AI performs literature searches to guide funding proposals AI evaluates literature based on multiple criteria, including which experts cite them. |
 | Moov – Operates a real-time digital marketplace for used semiconductor manufacturing equipment – tools that cost $1M+ new and are usable for 15-25 years but typically idle after just a few years. | Thin market: Semiconductor manufacturing equipment is highly specialized. Even with high demand, it is difficult to find suitable equipment on the second-hand market. | AI extracts specs from equipment documentation AI does smart matching, then hands problem to human tech support. |
 | Xometry – AI-powered marketplace matching custom manufacturing jobs (CNC machining, 3D printing, injection molding, sheet metal) with 7,000+ manufacturing facilities globally. | Thin market: Semiconductor manufacturing equipment is highly specialized. Even with high demand, it is difficult to find suitable equipment on the second-hand market. | AI builds quotes and lead time estimates from uploaded CAD files AI does supplier matching AI does continuous learning from processed transactions |
 | YantraLive – B2B marketplace for heavy construction/mining equipment parts (bulldozers, excavators, hydraulic pumps) with AI-powered part identification and prediction. | Thin market: Parts are difficult to source after OEM warranty expires, equipment models/versions change yearly, and aftermarket suppliers are fragmented. | AI extracts information from minimal input (part numbers, descriptions) ML provides predictive guidance in buying decisions |
 | QuestPair – Marketplace for new and certified refurbished laboratory equipment (sequencers, microscopes, spectrometers, bioreactors) targeting research labs and biotech startups. | Thin market: Scientific equipment buyers and sellers use different networks/channels, equipment is highly specialized, and quality concerns with used equipment require trust-building | AI handles complex technical specifications AI does two-sided verification (quality checks for equipment, vetting of buyers/sellers) |
These are just a few of the rapidly expanding set of relatively new companies that are targeting various types of thin markets … and it is only a cursory list. Nonetheless, it makes the point that thin markets exist and many entrepreneurs are using AI to make them real and actionable. For Deeperpoint and Cosolvent, the message is clear:
- Thin markets are real and represent big new opportunities
- AI is valuable, and probably essential, to prying open those markets
- These and other startups are independently applying the ideas that Cosolvent is designed to support
This makes the case that a building a generic, open sourced design is worth the effort. Instead of fighting their way alone through their chosen thin market, these startups could be working from a common playbook. That playbook would speed them through the development of their basic infrastructure and let them more quickly come to terms with the real moneymaking challenges in their chosen vertical.