GTM is among the hardest unsolved problems for AI-native companies. Most playbooks were built for a different era. AI-native products need AI-native commercial motions — and very few people have figured out what that actually looks like in practice. At our next Accel Founder Connect, Shekhar Kirani (@skirani), Partner at @Accel, sits down with two people who have cracked different parts of this problem: •⁠ ⁠Barathram A, Partner at McKinsey, leads Go-to-market work for AI and software companies, advising and partnering with frontier labs, hyperscalers, tech and AI-natives on how to build and scale commercial motions. •⁠ ⁠Varun Anand co-founded Clay, the GTM platform of choice for AI-native sales teams. He didn't just theorise about AI-native GTM. He had to figure it out while building the product itself. Together, they'll get into first-principles playbooks, org structures, and what global go-to-market actually looks like when you're building AI-native. Sign up to attend in person in Bengaluru: luma.com/AccelAIGTMStrategie… Sign up to attend virtually: luma.com/AccelAIGTMStrategie…

A few days ago, Shekhar Kirani (@skirani), Partner at Accel (@Accel), moderated a conversation focused on understanding what it actually takes to build products using AI. He was joined by Anand Arvind of Zenoti (@gozenoti), Dhimil Gosalia of BrowserStack (@browserstack) , Noel Curtis of BookMyShow (@bookmyshow_sup) , Ramesh Parthasarathy of DreamTeam, ⁠and ⁠Vivek Sharma of 1Wrk – operators who’ve spent years in the trenches building AI into real systems, products, and workflows. The conversation stayed grounded in execution: what breaks, what changes inside teams, and what starts to matter once AI moves from demos into production. Some observations that came up repeatedly: 1. How you build matters more than what you build. The companies that last are the ones treating feedback, notes, and lessons learned as part of the foundation, not extras to add later. 2. Writing things down is now part of the product. The prompts you write, the odd cases you catch, the rules you set — that's what teaches your AI how to behave. 3. Every mistake should teach the system something. When things break, the fix shouldn't just be a patch. It should change how the system thinks, so the same mistake doesn't happen twice. 4. Don't confuse usage with progress. A lot of people are using AI. Fewer are shipping things that actually work, hold up, and keep working under pressure.

For many B2B revenue teams, the core challenge isn’t a lack of tools, it’s unreliable data spread across too many systems. The typical response is to add more software. But more tools often create more silos, making the underlying data problem even harder to solve. By the time a sales rep is ready to engage, critical account context is fragmented across systems. Adding AI on top doesn’t fix this, it just accelerates bad decisions. Sprouts.ai goes a layer deeper. They’ve built a continuously refreshed intelligence layer that cleans, enriches, and unifies account data, so their AI revenue agents operate on signals teams can actually trust. These agents identify the right accounts, surface buying intent, map decision-makers, and execute outreach — directly within Salesforce and Microsoft Dynamics. Karan Chaudhry, Co-Founder and CEO, Kapil Chaudhry, Co-Founder and CTO, and Avinash Nagla, Co-Founder and COO, founded the company in 2023. In under two years, customers are already seeing 3x more qualified leads and 35% lower GTM costs. We’re continuing our partnership with the team as they reimagine how modern B2B revenue organizations operate. @skirani • #AccelFamily

We recently hosted a closed-door conversation with some of India's top consumer tech founders. This post continues the learnings we shared a few days back. Shekhar Kirani (@skirani), Partner at Accel, walked the room through the four stages every company moves through on the path to AI transformation. Most Indian companies are stuck somewhere between stage one and stage two. The four stages are in the carousel. A few patterns stood out across the companies furthest along: → One co-founder fully dedicated to the transformation, leading a team of 5 to 10 enthusiasts. The rest of the founding team runs the business. → Individual productivity and company-level automation in parallel. Pick one slice of the business, automate it end-to-end, use it as proof, then expand. → Hiring freezes unless the team can show AI is being used. @prashanthp • @binnybansal • @ankitnagori27 • @phanikishan

Credit awareness in India has grown sharply, with over 183 million people now actively tracking their scores, according to CIBIL. Most platforms are built to show you your score, but don’t always help consumers improve it. Oolka (@oolka_ai) is changing that. Its AI agents read your credit profile and act on it, filing disputes, closing dormant accounts, negotiating rates, and managing EMIs. It's the equivalent of having a personal relationship manager, available to every Indian at scale. Oolka was founded by Utkrishta Kumar (@youkaey), who spent years leading business at Meesho before starting up in 2024. We're partnering with Oolka as they build from credit health to a full financial companion for every Indian. @SaSarthaksingh • #AccelFamily

Retail investors in India have come a long way. Today, millions of people are actively trading options, building strategies, and making split-second decisions. ⁠But most platforms were built for first-time investors. For the serious, active trader, that gap is felt every single day. Jumping between tabs to check charts, confirm data, and place an order, while the market moves faster than the platform can keep up. Sahi (⁠@Sahi_HQ) was built to fix exactly that as an AI-powered broking platform designed from the ground up for active traders, where charting, analysis, and execution all happen on one screen. ⁠Behind this is a founder duo that understands both sides of the problem. Dale Vaz (@dale_vaz) brings hands-on trading experience and the engineering instincts of a former CTO of Swiggy. Manish Jain (@BeingMojo) built professional trading products at Kotak Securities and understands exactly where retail traders could be served better. ⁠We’re deepening our partnership with them as they continue to shorten the distance between insight and execution for traders. Manasi Shah • #AccelFamily

Prayank Swaroop (@prayanks), Partner at Accel, was at Off The Record last week, hosted by Venture Catalysts (@vcatsindia) and Seligman Ventures, and shared his honest take on enterprise sales in 2026. ⁠AI product buyers have sat through hundreds of demos. They're no longer asking if your product works; they're asking where the business impact is coming from. If you can't answer that fast, you're losing deals you'd have won 18 months ago. ⁠Don't lead with the big pitch. Start with one person who finds your product indispensable. Then a team. Then a department. By the time it reaches the decision-makers, the product has already made its case. That's the path that's actually been working.

Forbes’s AI 50 List highlights the world’s most promising privately-held AI companies, judged on commercial traction, technical talent, and how seriously they've embedded AI into what they build. @AnthropicAI, @cursor_ai, @cyera_io, @DecagonAI, @GammaApp, @WeAreLegora, @Lovable, @perplexity_ai, @speak, @SynthesiaIO, and @thinkymachines made the list. In what Forbes called a highly competitive field, this is a testament to these teams’ sharp ideas and real technical depth. Read the full article: forbes.com/lists/ai50/