SeedToScale | Curated by Accel

We’re in the early innings of AI, and the playbook for building companies is already changing.

In this episode of SeedToScale, Shekhar Kirani, Partner at Accel, sits down with Jason Graefe, CVP of Microsoft’s AI Partner Catalyst Team, to unpack what’s actually changing in the AI era.

From building products to winning enterprise customers, the playbook is being rewritten in real time.

Jason brings a unique vantage point, working closely with some of the fastest-scaling AI companies globally, while also operating inside Microsoft at the forefront of AI transformation. The insights are practical, grounded, and directly relevant to founders building today.

Here’s what we explore in this episode:

• Why we’re still in the “early innings” of AI, and what that means for founders

• The shift from horizontal AI tools to durable, workflow-driven products

• Why the first enterprise sale is always a trust sale

• How design partners can accelerate product-market fit

• Speed as a moat: why faster iteration is defining winners

• The role of “taste” in building products users actually stick with

• Why go-to-market needs to be built from day one, not as an afterthought

• The “golden age of margin”: reducing costs while creating entirely new products

• India’s structural advantage in building and scaling AI startups

If you’re a founder, builder, or operator trying to understand how to win in the AI cycle, this conversation is a clear, grounded playbook.

Chapters:

0:00 - Introduction

0:58 - India’s startup energy and why this market stands out

2:07 - What’s changed since pre-COVID

3:18 - Global AI ecosystems and where India fits

4:42 - The rise of second-generation founders in India

6:05 - Why startups are scaling faster than ever

7:26 - From Flipkart to AI-native companies: the infra shift

9:12 - Is the AI market already crowded?

10:36 - Why we’re still in the early innings of AI

11:48 - AI in traditional industries: where we actually are

12:31 - The “golden age of margin” explained

13:52 - Startups hitting profitability faster with AI

15:04 - Moving atoms vs moving bits: what AI can’t disrupt easily

17:18 - Where the real opportunities still exist

18:42 - Horizontal vs vertical AI: speed vs durability

20:11 - Building AI products that adapt to enterprises

22:07 - Why personalization creates sticky products

23:34 - Software that feels like it’s built for you

25:02 - Enterprise AI adoption: two buyer behaviors

26:41 - Why the first sale is always a trust sale

28:06 - How to win your first design partners

29:52 - Enterprises are more open than ever to co-building

31:08 - Surviving rapid LLM evolution

32:14 - Taste: the ultimate differentiator in AI products

33:27 - Real-time feedback loops → faster product iteration

34:52 - Speed as the new moat

36:09 - Domain experts vs first-principles builders

37:42 - First-mover advantage in AI

39:03 - How product + GTM cycles are compressing

40:18 - Why GTM must be built from day one

42:06 - Experimentation in go-to-market

43:38 - India as a proving ground for global products

45:11 - Why early customers matter more than big logos

47:02 - Product-led vs forward-deployed models

48:36 - Selling outcomes, not software

50:07 - India’s structural advantage in AI talent

51:19 - The culture of learning driving adoption

52:27 - Outro

Follow SeedToScale on LinkedIn: / seedtoscale

Follow Shekhar Kirani on LinkedIn: / kirani

Follow Jason Graefe on LinkedIn: / jason-graefe

#SeedToScale #Microsoft #Accel #AI #AIFounders #Startups #ArtificialIntelligence #Product #GoToMarket #IndiaTech #AIStartups

In this episode of PitchCraft, Ixigo MD and Group CEO Aloke Bajpai joins early investor Shailesh Lakhani to revisit the company’s 2016 Series B deck, a moment that captured Ixigo at an inflection point in India’s evolving travel market.

The conversation traces how Ixigo moved beyond being a travel meta-search engine by focusing relentlessly on solving customer pain points across flights, trains, and buses, while betting early on mobile growth and product-led innovation.

Shailesh reflects on what built conviction after years of tracking the founders: disciplined execution, resilience through multiple crises, and the willingness to pivot from aggregation to owning transactions when customer experience demanded it.

They unpack the strategic shifts, and long-term decisions that shaped Ixigo’s journey from startup to publicly listed travel technology leader.

Founders today are building in an environment shaped by constant technological shifts and evolving team dynamics.

In this episode, Gokul Rajaram, Partner at Marathon Management Partners and early backer of companies like CRED, Figma, Curefit, and Postman, shares how he thinks about building startups that endure. Drawing on his experience as a product leader, investor, and board member, Gokul reflects on the patterns he’s seen across hundreds of companies.

The conversation explores how founders navigate inflection points, scale teams responsibly, make irreversible decisions, and earn trust over time—with customers, teams, and investors. Gokul also speaks candidly about what tends to break early companies, and what quietly compounds when founders get the fundamentals right.

Behind every funding decision lies hours of evaluation founders never get to witness.

This episode of SeedToScale offers a rare backstage pass into the room where ideas are dissected, and where a simple yes or no is anything but simple.

Seasoned investors, Mukul Arora, Co-Managing Partner at Elevation Capital, and Rajan Anandan, Managing Director at Peak XV, sit for a discussion with Anand Daniel, Partner, Accel, to unpack what really happens, from the first cold email to the final investment committe vote.

The conversation goes beyond the cliches and gets brutally honest about how thousands of startups get filtered.

“Founders need to stop assuming VCs are Excel junkies.”

In this episode of PitchCraft, Pratik Agarwal, Partner at Accel, breaks down how investors really evaluate startups today.

We’re no longer in a copy-paste playbook era. Founders are building unfamiliar ideas across AI, consumer, SaaS, fintech, and deep tech and investors are comparing radically different businesses side by side. In that world, numbers alone don’t create conviction.

Pratik introduces the CORE framework, a universal pitch memo designed to help founders communicate what truly matters:

• Change: What has shifted in the world that makes this idea inevitable now

• Opportunity: Why this is a real, urgent problem with undeniable pull

• Resilience: How the business can compound, defend itself, and become a default

• Edge: The founder’s unique perspective and right to win

In this episode. Pratik walks through how this framework is already being used, and why it works across geographies, stages, and sectors.

If you’re a founder raising capital, refining your pitch deck, or struggling to get investors to really get your idea, this episode is your playbook.

Doctor visits are short. Health journeys are long.

In this episode, Anuruddh Mishra, Founder & CEO of August AI, talks about building an AI-led health companion designed to support patients beyond brief clinical interactions. He shares how August is built to retain personal health context over time, stay available when questions come up, and help people make sense of scattered medical information across their journey.

Watch the full episode for the complete conversation.

Chapters:

0:09 - Intro

0:33 - What is an AI health companion

3:06 - As an engineer, why did you choose to get into this field related to health?

8:04 - Did you worry about ChatGPT being a competition?

10:15 - How do you look at benchmarks?

14:31 - How is it performing versus human doctors?

15:40 - Why are you a horizontal health AI platform?

17:53 - Where does August AI stand in terms of memory?

21:37 - Is August’s memory different from general purpose memory?

21:53 - Are there any general purpose memory tools or do you have to build everything in-house?

23:00 - How much did you spend on marketing?

24:15 - What did you do in the early days to get this magical market pull?

27:30 - How did you go about fundraising & was it challenging?

33:30 - Aren’t you worried about Google & ChatGPT coming at August with full force once they take notice?

36:00 - Can you help us walk through your version of the world filled with an abundance of clinical knowledge?

38:54 - How do you ensure that AI never crosses regulatory guardrails?

42:36 - In 10 years what will be inevitable with AI in healthcare?

44:36 - Outro

Follow Anuruddh on LinkedIn: https://www.linkedin.com/in/anuruddhmishra/?originalSubdomain=in

Follow Anagh Prasad on LinkedIn: https://www.linkedin.com/in/anaghprasad/?originalSubdomain=in

Follow SeedToScale on LinkedIn: https://www.linkedin.com/company/seedtoscale/

#DecodingAI #SeedToScale #Accel #Healthcare #Tech #Medical #Startup #Funding #Digital

Everyone’s talking about generative AI.

Few talk about industrial AI: where the data meets the factory floor.

In this episode of Decoding Manufacturing, Vipin Raghavan, Co-founder & CEO, Haber, explains how Haber is using AI to close the loop between lab and line, turning process data into real-time chemistry control and measurable profit impact.

From automating dosing systems to building trust through prediction accuracy, Vipin breaks down what it takes to bring AI into legacy plants, why Haber prices by production output instead of licenses, and how vertical depth and reliability built a zero-churn business.

He also talks about India’s manufacturing readiness, cultural differences in adoption, and what the road to autonomous factories really looks like.

Chapters

00:00 – Intro: AI that augments, not replaces operators

02:57 – The three universal manufacturing challenges

07:25 – Why now: Connected factories and cheaper edge compute

09:59 – Trust before autonomy: Building accuracy first

13:52 – The pacemaker product and Haber’s zero churn

17:37 – Pricing by the tonne, not by the seat

21:46 – India vs West: Cultural adoption of industrial AI

24:36 – Building T-shaped talent

27:33 – The long road to near-autonomous factories

Follow SeedToScale on LinkedIn for more insights: https://www.linkedin.com/company/seedtoscale

Follow Vipin on LinkedIn: https://www.linkedin.com/in/vipin-raghavan-6142986/?originalSubdomain=in

#AI #Manufacturing #Automation #Innovation #Technology #Sustainability #Industry

#Data #DecodingManufacturing

Everyone’s talking about AI models. Few talk about the data that moves them forward.

In this episode of Decoding AI, Jonathan Siddharth, Founder & CEO, Turing, explains how Turing evolved from a global developer platform into a research accelerator, fueling the world’s top labs and enterprises with bespoke, interdisciplinary training data.

From RL gyms to post-training pipelines, Turing is engineering the human-in-the-loop systems that are quietly pushing us toward ASI.

Chapters

00:00 – Intro: The Future of Intelligence

01:11 – From Stanford to Rover: Jonathan’s Founder Journey

07:05 – The Idea Behind Turing

09:28 – Why Post-Training Is the Real Edge in AI

12:49 – How the World’s Smartest Humans Move Models Forward

19:10 – Inside Alan: Turing’s Data Engine & Talent Network

21:17 – How Enterprises Build Their AI Advantage

25:28 – The 100× Productivity Shift Ahead

29:59 – Closing Thoughts: The Future of Work & Intelligence

Follow Anand Daniel on LinkedIn: https://www.linkedin.com/in/ananddaniel/

Follow Jonathan on LinkedIn: https://www.linkedin.com/in/jonsid/

Follow SeedToScale on LinkedIn for more updates: linkedin.com/company/seedtoscale

#startup #ai #artificialintelligence #turing #data #ASI #artificialsuperintelligence #intelligence #aimodels

The next big shift in AI isn’t about models getting smarter, it’s about how we work.

In this episode of Decoding AI, Aparna Chennapragada, Chief Product Officer for AI Experiences at Microsoft, shares how AI is transforming the workplace from tools that assist us to teammates that collaborate with us.

Aparna draws from over two decades of product building. From Akamai’s internet backbone to Google’s AI-first products, like Lens, and now leading Microsoft’s AI strategy. Her insights go beyond features and hype, focusing instead on what it takes to build AI products that last.

Here’s what we explore in this episode:

• Why “adding AI” isn’t enough and what “AI-first” really means

• How Microsoft is reimagining productivity for the AI era

• What happens when the model starts to “eat the product”

• Why trust, timing, and design are key to durable AI

• Where founders can find the next big opportunities in AI, from context engineering to workflow intelligence

If you’re a founder, builder, or product leader trying to understand where AI is headed next, this conversation is a roadmap.

Chapters:

00:00 – Introduction

01:32 – Aparna’s journey: From Akamai to Google to Microsoft

05:20 – Lessons from building early AI products at Google

08:15 – Why the timing of AI adoption matters more than novelty

11:40 – The speed of AI change

15:05 – The shift from tools to teammates at Microsoft

19:30 – How AI is transforming productivity inside organisations

23:10 – The danger of “adding AI” vs. building AI-first

28:20 – What happens when the model eats the product

33:05 – Building moats: data, context, and trust

37:25 – AI products that last beyond hype

41:00 – Where founders can build next in AI

45:12 – Closing reflections

Follow SeedToScale on LinkedIn: https://www.linkedin.com/company/seedtoscale/

Follow Anand Daniel on LinkedIn: https://www.linkedin.com/in/ananddaniel/

Follow Aparna on LinkedIn: https://www.linkedin.com/in/aparnacd/

Follow SeedToScale on X: https://x.com/Seed2Scale

#DecodingAI #SeedToScale #Microsoft #Accel #AI #ProductDesign #FutureOfWork #startup #aistartups #artificialintelligence #copilot #accel

Every metal behaves predictably, but carbon fiber doesn’t.

Its strength depends on how every fiber is laid, making it one of the hardest materials to automate.

That’s where Dhinesh Kanagaraj, Founder & CEO, Fabheads, comes in. Their breakthrough “adaptive tow placement” technology lets machines think like craftsmen, achieving fiber-level control at industrial speed.

With software-driven automation and smarter material handling, Fabheads is achieving 3-10X productivity, up to 40% fewer rejections, and reduced waste.

Watch this episode of Decoding Manufacturing to see how Fabheads is transforming carbon fiber manufacturing.

Chapters:

0:24 - Intro

0:28 - Revolutionizing the carbon fiber industry

1:05 - Misconception about automating carbon fiber manufacturing

1:32 - The single most important breakthrough in tech stack

2:07 - Brief explanation of what Fabheds does

2:50 - Misconception about automating carbon fiber manufacturing

3:27 - The single most important breakthrough in tech stack

4:00 - Solving the gap in India’s composite manufacturing ecosystem

5:15 - Carbon fiber 3D printing for aerospace

7:41 - Demand for robotic composite system in a manual manufacturing led industry

8:21 - Elaborate on the tech stack

9:33 - The hardest technical milestone

10:17 - Comparison of robotic automation to traditional manual work

11:05 - The best yet underappreciated tech stack

12:13 - Reasons behind building end-to-end approach

13:52 - What keeps high profile clients engaged to Fabheads

10:39 - The economic structure of Fabheads

15:21 - Biggest growth challenge

16:43 - Operational challenges with expansion

17:58 - Building talent in a niche ecosystem

19:10 - Failures and early lessons

20:36 - Vertical vs horizontal Industries

22:17 - One belief that needed unlearning

23:22 - Success for Fabheads in 5 years down the line

24:06 - Outro

Follow SeedToScale on LinkedIn for more insights: https://in.linkedin.com/company/seedt...

Follow Dhinesh on LinkedIn: https://www.linkedin.com/in/dhineshrk/?originalSubdomain=in

#DecodingManufacturing #CarbonFiber #Automation #Engineering #Innovation #DeepTech #Composites #SeedToScale