Floqer to Open-Source Its AI Customer Database in Early 2027

Floqer plans to open-source the technology behind its autonomous customer knowledge database in early 2027, pending security reviews. The company disclosed the plan to TechCrunch ahead of its pitch on Wednesday on the Startup Battlefield stage at TechCrunch Disrupt 2026 TechCrunch.

Disrupt runs October 13 to 15 at Moscone West in San Francisco. Floqer was named to the 2026 Startup Battlefield Top 20, a cohort drawn from a process whose application deadline was June 8 and whose 200 competing startups were chosen in August.

Floqer describes itself as an autonomous customer knowledge platform for B2B sales and marketing teams. The core is an AI-powered database to collect, organize and update customer information, sold as AI data enrichment for CRM, the system sales teams use as their record of contacts and accounts. The company was founded by CEO Shivam Mahajan and Zaaheda Islam, who met at university. Mahajan previously worked as a software engineer at Kyndryl, and Islam previously worked as a senior data scientist at Axis Capital.

Its first product was different. The team built a Chrome extension called Super Chat that let users log in with a ChatGPT account to understand information and take notes. That experiment in retrieval and note-taking grew into the current platform for prospecting and account research. Users start with natural language to describe what they are looking for, and Floqer agents search across sources to build lists of relevant companies and contacts. For enrichment, Floqer says its system draws on more than 100 third-party data sources, up from the 80-plus figure used in the Battlefield preview list TechCrunch.

For contact data, the company uses a waterfall approach, meaning it checks one data provider after another until it finds usable information for verified emails and phone numbers. The company states the waterfall selects the best match per field and charges only for hits. Every email is bounce-checked and every phone number validated before it reaches the customer CRM. On top of structured vendors, Floqer operates AI web research agents to find answers when no database has them, with a source provided for every result. It states it tracks more than 80 signals including funding, hiring, website visits, search intent, G2 reviews and Reddit mentions. It also claims detection of job changes, dead emails and dead domains the day they happen. The agents are said to deliver accurate emails, phone numbers and more than 100 data points per customer record.

Commercial traction is early. Floqer says it has surpassed $1.2 million in annual recurring revenue, the subscription revenue counted on a yearly basis, with overall revenue growing roughly tenfold over the past year. It states it has cleaned and enriched more than 10 million records. Customers have reported cutting prospecting time by as much as 70%, according to the company. Two deployments are named. Floqer lists Perplexity as a customer case study, with reference to Head of Enterprise Ops and Systems Nate Follen on enriching every signup. It lists Pulumi as a customer, stating its agents read repos across more than 40,000 CRM accounts to flag accounts pushing infrastructure code.

For enterprise buyers, Floqer states it is SOC 2 Type 2 certified, an independent audit of security controls, and GDPR compliant, meeting European privacy rules. It states it never uses customer data to train AI models and never shares or sells customer information to third parties. On its Enterprise plan, data deleted in Floqer is permanently erased after seven days.

The broader context here is that sales data decays fast. Titles change, domains die, people move, buying signals fade. Floqer is betting that agents that continuously research, verify and rewrite records can keep the system of record current without human list building. In my view, if the open-source release in early 2027 clears security review, it will test whether data quality in sales tools improves when the infrastructure is shared or whether the edge stays with exclusive sources and private graphs. Either result gives practitioners a clearer basis for choosing where to place their CRM automation bets.