Orange Slice Is a Spreadsheet That Writes Its Own TypeScript and I Am Into It

The Macro: Sales Enrichment Is a Mess of Tabs and API Keys

I have worked with enough B2B sales teams to know what the enrichment workflow actually looks like in practice. It is not pretty. Someone downloads a list from LinkedIn Sales Navigator. Someone else cross-references it with ZoomInfo for email addresses. A third person runs those emails through a verification tool. Then somebody opens Crunchbase to check funding status, BuiltWith for tech stack data, and maybe a news aggregator to see if the company just raised or laid off half its team. All of this happens across six browser tabs, three different logins, and a Google Sheet held together with VLOOKUP formulas and prayers.

The sales enrichment market is big and fragmented. ZoomInfo charges enterprise prices for a database that is stale by the time you export it. Apollo.io is cheaper but thinner. Clearbit (now Breeze by HubSpot) is decent for firmographics but weak on contacts. Lusha focuses on direct dials. Clay built a workflow tool that connects these sources together, which was a step forward, but it still requires manual configuration and the learning curve is steep. Cognism, LeadIQ, Seamless.AI, they all cover some subset of the problem.

The fundamental issue is that no single data provider has everything, and stitching together multiple providers is an operational nightmare. You end up paying for three or four tools that each give you 60 percent of what you need, with overlap in the middle and gaps at the edges.

What is interesting about the current moment is that AI makes it possible to build enrichment workflows dynamically instead of configuring them statically. Rather than setting up a fixed pipeline that pulls from Source A, then Source B, then Source C, you can describe what you want in plain language and let the system figure out the execution. That is a fundamentally different approach, and it is where the category is heading whether the incumbents like it or not.

The Micro: A Nineteen-Year-Old Exit and a Broadway Choreography App

Orange Slice is a spreadsheet where every column runs TypeScript code generated by AI. You describe the data you want in a chat interface, and the system writes the code, creates the columns, and executes across your entire list. It pulls from over 100 built-in data sources including LinkedIn, company databases, web scrapers, and AI research agents. The platform handles dependency ordering automatically, so if Column C needs the output of Column B, it runs them in the right sequence.

Vihaar Nandigala is CEO and Kishan Sripada is CTO. They are a two-person team out of San Francisco, part of Y Combinator’s Summer 2025 batch. Vihaar sold his previous company at 19 and worked at J.P. Morgan. Kishan built FORMI, a choreography collaboration platform that was adopted by Broadway, Disney, and the NBA, reaching over 100,000 users. He also interned at Ramp and worked at Tour, a YC S21 company. These are builders who have shipped real products to real users, which is more than I can say for a lot of teams in the sales tech space.

The TypeScript angle is what makes Orange Slice genuinely different from Clay or the other workflow tools. Clay uses a visual builder with pre-built integrations. Orange Slice generates code. That means the enrichment logic is transparent, editable, and extensible. If the AI-generated code does not do exactly what you want, you can modify it. If you need a custom enrichment that pulls from your company’s internal API, you can write that column yourself. The system is not a black box.

They have raised $5.3 million in seed funding and claim over 5,000 sales and RevOps teams as customers, with logos including Oracle on the website. For a company that recently came out of YC, those numbers suggest fast traction. The product also includes a CRM agent that lets you query Salesforce or HubSpot conversationally, and a social listening agent that monitors Reddit, LinkedIn, and X for buying signals. That is a lot of surface area for a two-person team, which could be either ambitious or overextended.

The Verdict

Orange Slice is betting that the future of sales enrichment is code, not configuration. I think that bet is right. The visual workflow approach that Clay pioneered is better than doing everything manually, but it hits a ceiling when you need custom logic or unusual data sources. TypeScript columns that the AI writes for you combine the accessibility of a spreadsheet with the power of a real programming environment.

The risk is complexity management. Spreadsheets that run code across thousands of rows are powerful until something breaks, and then debugging a TypeScript column that an AI wrote for you is a worse experience than debugging one you wrote yourself. Error handling, rate limiting on external APIs, and data quality checks all need to be bulletproof or the product becomes a liability for the sales ops person who adopted it.

In thirty days, I want to see how sticky the product is. Are sales teams using it daily or did they run one enrichment and go back to ZoomInfo? Sixty days, the question is whether the TypeScript transparency matters in practice. Are users actually editing the generated code, or is it a feature that sounds good in a demo but never gets used? Ninety days, I want to see how they handle the 5,000-customer base with a two-person team. Customer success at that scale requires either extraordinary product quality or a support team they have not hired yet. The product concept is the most interesting thing I have seen in sales enrichment since Clay launched. Whether the execution matches the concept is the open question.