Signals That Matter: Venture Capital Evaluation of Early-Stage AI Fintech Startups
Advisor: Mr. Koppel
Internship Location: Bright Ventures
Onsite Mentor: Lenore Champagne-Beirne, Managing Partner
Artificial intelligence is rapidly changing the startup landscape, raising new questions about how venture capital investors evaluate early-stage companies. This project investigates how the venture capital due-diligence process has evolved as artificial intelligence becomes more widely integrated into startups. While venture capital has traditionally focused on factors such as market demand, scalability, competition, and growth potential, AI introduces additional considerations including proprietary data, system scalability, and potential risks associated with automated decision-making. This research is conducted in collaboration with Bright Ventures, a New York–based venture capital firm focused on early-stage startups. The project combines analysis of venture capital with direct observation of investor meetings, startup evaluations, and a survey distributed to venture capital professionals. These methods allow for a comparison between evaluation frameworks used before the rise of AI and those used today. The project expects to show that although traditional investment fundamentals—such as customer demand – remain essential, AI expands the due-diligence process by introducing new technical, ethical, and scalability considerations. Understanding how these evaluation frameworks are evolving helps clarify how investors identify startups with long-term potential in an increasingly AI-driven economy.
Project Blog
Week 0: Introduction
Hello everyone, my name is Saajan Parekh, and welcome to my Week 0 Senior Project Blog! Over the next ten weeks, I will be researching how artificial intelligence is changing the way venture capital investors evaluate startups. In this blog, I will introduce my research topic, explain why it matters, and outline what I plan to accomplish in the first week of my project.
Importance of the Subject
Venture capital plays a major role in shaping innovation. Because many startups fail, investors must carefully analyze companies before investing. This process is known as due diligence, where investors evaluate factors such as the founding team, market opportunity, customer demand, and financial projections. Traditionally, due diligence relied heavily on manual research, founder interviews, and market analysis. However, the rapid rise of artificial intelligence is beginning to change how this process works. Investors can now analyze large datasets and evaluate user behavior much faster than before. At the same time, the increased use of AI raises new questions about how venture capital firms interpret data, assess risk, and determine which companies truly have long-term potential.
For this reason, my research question is:
How does AI affect the due-diligence process for startups, and how has it changed relative to prior processes before AI dominated the analysis?
Research Project and Advisors
To explore this question, I will be working closely with professionals in the venture capital industry at a firm named Bright Ventures. My project will involve learning about how venture capital firms evaluate early-stage startups and how those evaluation methods have evolved as AI tools become more common in the investment process. Additionally, I will gather insights through interviews and research on venture capital evaluation frameworks through a survey that I will send to multiple professionals using Bright Ventures connections.
Week 1 Future Plan
During Week 1, I plan to begin developing a framework for analyzing how venture capital firms perform due diligence. Specifically, I will:
Research how venture capital due diligence worked before widespread AI adoption
Identify key evaluation criteria investors use when assessing startups
By building this framework early in the project, I hope to shape my understanding of how AI tools are reshaping investment decisions. I am excited to begin this research and share my progress in the coming weeks. Thank you for reading, and I look forward to updating you in Week 1!
For those of you who want to view my on-site placement company information, here is the link to their website: https://www.brightventures.io/
For those of you who want to read over my original proposal for this senior project, here is the link to the google doc: https://docs.google.com/document/d/1NF_aQtRqQuY7pa9DqDAviE6WMsPyzAEY6EBb3JytlYE/edit?usp=sharing
Week 1: The Start
Quick recap of Week 0 before I get into it: I introduced my research question: How does AI affect the due-diligence process for startups, and how has it changed relative to prior processes before AI dominated the analysis? I also laid out my plan to explore it through my internship at Bright Ventures, an early-stage VC firm in New York.
My whole week ended up being virtual because of some scheduling conflicts, but I still had work to do. It started with me sitting in on calls for a few of Bright Ventures' portfolio companies. Before the calls, I went through the investment memos and pitch decks for each company to get some background, jotting down important information on them. I also tried to connect everything back to my research question the whole time, thinking about which parts of that process AI might be changing. Later in the week I started working on my investor survey, which is going to be one of my main research tools. I hope to send it out to VC professionals through Bright Ventures' network, so I spent a lot of time making sure the questions actually flow well and build on each other — starting with how due diligence worked before AI, then moving into how it works now, and ending with some open reflection questions.
At the end of my week, I had a check-in meeting with my on-site mentors. I went in thinking AI's biggest impact on due diligence was just making things faster, but they explained that Bright now evaluates startups on three AI-specific layers: how AI is changing the market the startup is in, whether the product is actually AI-native, and whether the technology handles data responsibly. AI has to be additive; It needs to genuinely help users, not just extract value from them. Most notably, because AI moves so much faster and at a bigger scale than past technologies, the cost of a wrong investment decision is way higher now than it used to be. Everything I learned from this week will definitely help to contribute to my research for this project.
Next week, I will be on-site, so I will have more information to go off of from there. I hope to understand more about the due-diligence process next week through the work Bright Ventures gives me! More updates in the coming weeks!
Week 2: First Week On Site
Quick recap of last week: my first week was fully virtual, but I still sat in on portfolio company calls, reviewed investment memos, and had a check-in with my mentors that reshaped how I think about AI's role in the due diligence process. This week, I went on-site at Bright Ventures.
I spent the majority of my week building out a research spreadsheet of early-stage startups across Bright Ventures' core investment verticals: fintech, healthcare, and future of work. For each company, I looked at what they do, what stage they're at, why they fit Bright's investment mandate, and how AI plays into their product. I did a lot of research, but it ended up being one of the more eye-opening parts of the project so far.
What stood out to me was that every single one of the companies was using AI to make an existing field meaningfully better. Whether it was automating legal work, improving patient monitoring, or streamlining financial processes, the pattern was the same: AI wasn't the product, it was what made the product actually work. That distinction is very important to my research question.
On Tuesday, I went to Bright Ventures in person and got to sit down with Lenore and her Chief of Staff, Ciera, to walk through the spreadsheet, talk through my senior project progress, and map out next steps. The meeting was mostly logistical, but getting that time in person helped me make sure my work aligns with what Bright is looking for.
I've also been continuing to refine my investor survey, which I'm planning to send out in about two weeks. I want to make sure the questions are sharp before it goes out, since the responses will drive one of the most important parts of my research.
Next week I'll be continuing on-site, and I'm hoping to dig deeper into Bright's actual evaluation framework. More updates soon!
Week 3: Building the Framework
Quick recap of last week: I went on-site for the first time, built out a research spreadsheet of early-stage startups across Bright Ventures' three core verticals, and continued refining my investor survey. This week, I dived into my biggest project.
This week I was on-site at Bright Ventures working on what I'd say is the core deliverable of my project so far: a side-by-side comparison framework of traditional due diligence versus AI-native due diligence. The goal is to map out how the questions VCs ask have shifted now that so many startups are built on AI models, APIs, and proprietary datasets.
To build it, I organized the framework into five categories: Product & Technology, Team & Talent, Market & Competitive Moat, Business Model & Unit Economics, and Risk. For each one, I researched what investors asked before AI-native companies dominated deal flow, what they ask now, and why the frameworks has changed. I did my research and through that, I built my framework based off articles I found.
For my sources, I pulled from public writings by firms like a16z, Sequoia, and First Round Capital. A lot of their work touches on questions like whether a startup has a real AI moat, whether the product is AI-native (The AI is adding value to the startup) or an AI wrapper (the AI is built on top of an already existing LLM, essentially making it so that they don't own any proprietary data). This turned out to be the important angles for filling out the framework.
The biggest takeaway so far is that the core goals of due diligence haven't changed; Investors still want to know if a company can build something scalable in a big market with the right team, but the way you answer those questions looks completely different when the product is built on AI. This gap is exactly what my research question is trying to capture.
Next week I'll finish up the framework and write a short narrative summary of the most important shifts. More updates soon!
Week 4: Adjusting my Focus
Quick recap of last week: I was on-site at Bright Ventures building a side-by-side comparison framework of traditional due diligence versus AI-native due diligence, organized across five categories — Product & Technology, Team & Talent, Market & Competitive Moat, Business Model & Unit Economics, and Risk. This week, I wrapped that project up.
I finished the comparison framework and completed the one-page narrative summary that goes along with it. The narrative focuses on the most important shifts — specifically how the risks that were considered in traditional due diligence have changed with that of AI-native due diligence, as well as the type of competitive advantage that traditional startups had versus the competitive advantages an AI-native startups needs to have. That felt like a good note to close the project on.
The other big update this week is about my investor survey. Originally, I was hoping Bright Ventures could provide a list of contacts to send it to. That didn't work out, so I'm adjusting my approach. Next week I'll be researching investors on LinkedIn and looking into venture capital firms around the country to build my own contact list. It's a bit more work, but it also means I'll have a broader range of respondents beyond just Bright's immediate network, which could make my data more representative.
The survey itself still needs a few updates before it goes out, so I'll be refining the questions next week. More updates soon!
Week 5: Sending the Survey
Quick recap of last week: I finished the traditional vs. AI-native due diligence comparison framework and its narrative summary, and I started planning my outreach strategy after learning Bright Ventures couldn't provide a contact list for my survey.
This week I was on-site at Bright Ventures, and the main focus was getting my survey out into the world. I spent time reaching out to around 30 to 40 investors on LinkedIn, sending connection requests with the hope that they accept and I can follow up with the survey in the coming weeks. It was a lot of outreach, but building that contact list helps me pull from firms across the country, meaning I'll get a wider range of perspectives than I would have from a single network.
Alongside the outreach, I finalized my survey questions. After completing the comparison framework last week, I had a much clearer sense of what I actually needed to ask, so I went back and tightened everything up. The survey now feels like a solid research tool rather than a work in progress.
The responses I get back will be one of the most valuable parts of this project. Hearing directly from investors and founders about how their evaluation process has changed because of AI will give me real data to work with on top of my framework. I'm hoping the next few weeks bring in enough responses to build something meaningful.
Next week I'll be following up on connections and continuing on-site work. More updates soon!
Week 6: Still Waiting For Survey Results
Quick recap of last week: I reached out to around 50 investors on LinkedIn and finalized my survey questions after wrapping up the due diligence comparison framework.
This week I was on-site at Bright Ventures, and I started pulling everything together. Using the research, frameworks, and notes I've built up over the past several weeks, I began outlining my final presentation and research paper.
The one challenge this week is that I haven't received any survey responses yet. It has been quite an obstacle, since the survey was meant to be a key source of primary research for this project. I'm not giving up on it — the LinkedIn connections are still pending and responses could still come in — but I'm also starting to think through alternative ways to gather that data. I want to make sure my research has a strong foundation regardless of how the survey performs.
On that note, Bright Ventures is looking into setting up interviews for me with investors where I can ask my survey questions directly and potentially quote them in my research. Quotes from investors would add a new level of depth to my research.
I also reached out to Ms. Goldfarb, who offered to feature my senior project in the school newsletter. Hopefully that gets my survey in front of some parents who work in venture capital or finance and can contribute responses.
Next week I'll continue building out the presentation and paper while figuring out the best path forward for primary research. More updates soon!
Week 7: Compiling Research
Quick recap of last week: I started outlining my final presentation and research paper, reached out to Ms. Goldfarb about featuring my project in the school newsletter, and learned that Bright Ventures is looking into setting up investor interviews for me.
This week I was on-site at Bright Ventures, and honestly it was one of the more encouraging weeks of the project. I received three responses on my investor survey, which is more than I expected at this stage. The survey is completely anonymous, but the answers have already given me a lot of valuable information to work with. Hearing information about my project directly from people in the venture capital industry adds a different layer to my research.
On top of that, Ms. Goldfarb put my survey out in the school newsletter this week. I'm hoping that reaches some parents in the venture capital world who can contribute responses over the coming weeks.
I also continued building out my final presentation and research paper. Both are starting to come together well. I've been drawing on everything I've done over the past several weeks: the startup tracker, the due diligence comparison framework, and now the survey results. It's been very nice seeing how all of my work is able to be easily translated into a final presentation and paper.
I'm still waiting on more responses to come in, and I'm hoping the investor interviews Bright is setting up will add even more depth. The more primary research I can get, the stronger the final product will be.
Next week I'll keep pushing on the presentation and paper. More updates soon!
Week 8: Virtual and Lighter Week
Quick recap of last week: I received three survey responses, continued building my final presentation and research paper, and Ms. Goldfarb put my survey out in the school newsletter.
This week was virtual and a lighter week compared to the rest of the project. I kept my focus on the final presentation and research paper. Both are progressing well. I've been weaving in everything I've gathered over the course of this project — the startup tracker, the due diligence comparison framework, the survey results, and insights from my time on-site at Bright Ventures — and it's starting to feel like a complete, cohesive body of work.
One thing I've enjoyed about this stage is seeing how each piece of research connects to the next. The startup tracker gave me a ground-level view of what AI-native companies actually look like across different verticals. The due diligence framework gave me a structured way to compare how investors evaluated those companies before and after AI became a dominant force. The survey results are starting to add real investor voices to what was otherwise a research-based argument. Putting it all together has made the project feel much more substantial than any single piece would on its own.
I also now have five survey responses, which is encouraging. Every additional response adds more depth to my primary research, so I'm still hoping more come in before I finalize everything.
With the project wrapping up in the coming weeks, the priority right now is making sure everything is polished and well-supported. The research I've done gives me a solid foundation, and I'm focused on presenting it in a way that clearly answers my research question.
More updates soon!
Weeks 9-10: Working Towards the End Goal
I missed last week's post, so this entry covers both Week 9 and Week 10.
Week 9
Week 9 was virtual since the Bright team was traveling, so I worked independently on my final presentation. I also had a meeting with Ciera, Bright's Chief of Staff, to talk through how I could better use my survey results in the presentation. Since the response count was lower than I originally hoped, we discussed how to convey the information I did collect in the most specific and meaningful way possible. It was a helpful conversation and I realized that sometimes a smaller dataset forces you to be more precise and intentional with how you present your findings, rather than relying on volume alone. Alongside that, I spent time preparing for my dry run and continuing to work on my final paper, so that by the time Week 10 came around I felt ready.
Week 10
Week 10 was my last week with Bright Ventures. I did a dry run of my final presentation with the team, and the feedback was encouraging. They told me I presented well, and their main suggestion was to add more visuals to help the audience follow along. A lot of what I'm presenting involves frameworks and comparisons that are much easier to understand when you can see them rather than just hear about them.
I also continued working on my research paper and am very close to finishing it. Both the presentation and the paper just need some finishing touches before they are ready.
Looking back on the past ten weeks, this experience has shaped how I think about venture capital and AI. I came in with a research question and left with an answer, built from frameworks I developed, startups I researched, investors I surveyed, and conversations I had with professionals who do this work every day. I'm grateful to everyone at Bright Ventures for the opportunity and for pushing my thinking throughout the process.
More updates to come as I finalize everything!
Post 11: Conclusion Blog: Signals That Matter
When I started this project back in March, I wasn't sure exactly what I was getting into. I knew I was interested in venture capital and AI, but I didn't have a clear sense of what ten weeks of real research would actually look like.
My final presentation went really well. I felt confident delivering it, and the audience found the material interesting — which meant a lot given that this was a non-finance crowd. Getting people who aren't in venture capital to care about due diligence frameworks and AI evaluation criteria is its own challenge, and I'm glad it landed the way it did.
The core conclusion I reached through this research is something I didn't fully appreciate when I started: AI doesn't replace the fundamentals of due diligence, it adds an entirely new layer on top of them. Investors still care about the same things they always have: the team, the market, the business model, the product. But now they also have to ask whether the product is genuinely AI-native or just an AI wrapper, whether the data is handled responsibly, and whether the company can survive in a market where foundation models like ChatGPT are constantly expanding. Those are brand new questions that didn't exist five years ago, and they sit on top of everything else rather than replacing it.
This project pushed me in ways I didn't expect. Building frameworks, reaching out to investors cold on LinkedIn, refining a survey, doing a dry run at an actual VC firm all helped me realize the importance of my project. I'm walking away with a much deeper understanding of venture capital and an appreciation for how much rigor goes into evaluating a single startup. I'm grateful to everyone at Bright Ventures, especially Lenore and Ciera, for making this experience what it was.