🔥 Here's the Modern VC Tech Stack

Majority of modern VC funds run on the same tools. The 2026 DDVC Landscape contains unique insights from 345 funds, including their most frequently used vendors, where adoption of AI is the most mature & where it’s lacking.

Sourcing, screening, and DD score 3.4 out of 5 on our AI adoption scale. Engineering and infrastructure sit close behind at 3.3. Legal and compliance sits at 1.8, the lowest score on the value chain.

Interestingly, at the largest funds, that gap gets wider. Indication that the bigger the teams, the higher the inertia to get going.

Our Landscape report grouped the tools VCs actually use into five clusters.

Data. Harmonic, Findem, Evertrace, PitchBook, Specter, Coresignal, Crunchbase, Dealroom, CB Insights : the sourcing and intelligence layer VCs lean on to track companies and talent.

CRM, fund, and portfolio management. Vestberry, Carta, Affinity, Attio, Visible, Pipedrive, HubSpot, Salesforce, Rundit: the backbone for managing deal flow, cap tables, and LP/portfolio reporting.

Productivity. Slack, Granola, Wispr, Gamma, Notion, Airtable: how teams communicate, run meetings, and capture notes day to day.

Agents and automations. Kruncher, Claude, Cursor, ChatGPT, n8n, Langdock, Zapier, Perplexity, Gemini: the most valuable layer of the stack, where VCs are implementing AI for research and workflows.

Infrastructure. Foresight, Exa, Supabase, GitHub, Pinecone, PhantomBuster, Cloudflare, Google BigQuery, AWS: the systems powering search, data, and hosting.

Most of this list is now a commodity purchase. Any firm with a budget and the ability to cut through the noise can stand up a comparable stack in no time.

I've tracked VC tools since 2017, when the list I kept ran about 100 deep. It passed 1,000+ tools in 2025. My own working stack went the other direction.

I consolidated it from the peak of 80+ tools in 2024 down to about 30 tools in January this year, and most of them map straight onto the five clusters above: Affinity, Vestberry, and Carta for CRM; Harmonic, Dealroom, and Evertrace for data; Claude, ChatGPT and Gemini as the assistants; n8n and Zapier running the automation layer.

Full deep dive on my top 30 tools for 2026 here.

If everyone uses the same tools, is there still a way to generate alpha? My clear answer is yes - via actual adoption (don’t underestimate how difficult it is to change habits…), custom workflows, and most critically by fusing public data with proprietary workflow and decision data.

While we’ve covered the public vs private data part in detail in the past, the Landscape report went deep into custom workflows and AI/tool adoption. In the chart below, you can see it split by function, on a scale of 1 (=low) to 5 (=high).

Two functions clear 3.0. Everything else sits in the middle, and legal and compliance sits well below it.

The AUM breakout sharpens the picture. At firms managing $1 to 5 billion, engineering and product adoption climbs to 3.9, the highest score in the entire dataset. Legal and compliance at that same cohort drops to 1.5, the lowest. A 2.4-point gap between the two functions.

The pattern holds across the whole table: adoption tracks how verifiable and low-liability the work is, more than it tracks budget or firm size.

Vendors build fastest where the data is public or the workflow is already software, so sourcing, screening, and engineering are where the tool explosion actually landed.

Sourcing also has the lowest barrier of any function: the data is public, easy to check, and reviewed before anyone acts on it.

Sourcing turning into infrastructure instead of labor was the thesis we discussed previously in “Do You Still Need Analysts”. This year's Landscape data is the first real evidence firms are acting on it.

Engineering, or more specifically the builder stack within investment firms, leads for a similar reason. Code already gets tested and reviewed before it ships. Coding agents allow existing engineers to accelerate (=Fullstack VCs) and non-technical investors (=Workflow VCs) to finally build what wasn’t possible before.

CRM, fund ops, portfolio intelligence, and IR all run on mature point solutions. Affinity, Vestberry, and Carta, are proven products.

Adoption of AI in this part of the stack however still sits in the 2.1 to 2.6 range.

The tools work. The friction is not as high. Habits take longer to rebuild. Most of this tier also runs on seat-based pricing, not usage-based agentic spend. That keeps the cost predictable.

Overall the question remains “why change a running system?” - until it doesn’t. A reason why some of the top vendors proactively started pushing the frontier, like Affinity with their agent platform Affinity Ascend.

A score of 1.8 puts legal and compliance last, by a wide margin, on every cut of this data. At the largest funds, the gap gets worse.

I wrote about the actual mechanics of fixing this in "Compliance Is Where AI Co-Pilots Go To Die." The short version: most rollouts stall on five concerns nobody writes down. MNPI leaking into a model. LP data that legally can't touch a third-party system. No audit trail. Vendor risk. Access nobody can account for.

Name them and answer them in order. Compliance stops blocking the build and starts writing the guardrails alongside it. Low-risk data first, then read-only systems, then the firm's own playbooks encoded as skills.

Nothing in that sequence has changed. What the Landscape data adds is a clearer read on how large the gap actually is.

Every serious firm now has access to the same data providers, the same CRM, the same models. While the unique blend of vendors granted alpha for many years, it has become sort of a negative alpha to not have this access today.

What separates firms now is whether their own judgment, in investing and in legal work, has been turned into something a system can run under a governance model compliance actually trusts.

Legal and compliance sits furthest from that today, at 1.8 out of 5, worse still at the firms with the most resources to fix it. Fixing it turns compliance into the permission slip that lets a firm point AI at the data that actually matters: proprietary deal flow, portfolio signal, internal knowledge nobody can buy off a vendor.

That's the actual playbook, and running it is what closes the gap.

Stay driven,
Andre

PS: Join the DDVC Investor Summit at Bits & Pretzels virtually or physically in Munich during Oktoberfest Sep 28-30