Investor
ArchetypeCheck size: No person-specific or current firm check size is publicly disclosed. Archetype describes itself as an early-stage fund and its historical firm overview says it partners from first check through Series A; founders should confirm round fit directly rather than rely on directory estimates.
Aadharsh invests where applied cryptography makes useful software private, programmable and verifiable, and where decentralized AI can coordinate agents, data and computation without recreating centralized control. His Privacy 2.0 thesis centers on composable private shared state, hybrid PET stacks and eventually a broadly usable cryptographic protocol with explicit trust assumptions.
Technical teams that turn cryptography into fast, usable infrastructure; systems whose threat model and trust roots are explicit and independently verifiable; privacy that is composable and private by default; founders combining PETs around their strongest properties; programmable software that can hold value and act autonomously; and application-specific infrastructure intentionally designed around a real user culture.
His writing argues against isolated Privacy 1.0 tools with poor UX and no composability, closed hardware or supply-chain trust that users cannot reason about, one-size-fits-all claims that ignore threat-model tradeoffs, and generic chains optimized around abstract ideology rather than the application's users. No separate categorical pass list is public.
Reach him through the exact @oddhash account or LinkedIn linked by Archetype's live profile. Explain the concrete application, which data or computation must remain private, the adversary and trust assumptions, why the chosen mix of TEE/MPC/FHE/ZK is superior, performance and developer-experience evidence, and how the product becomes a reusable standard rather than an isolated privacy silo. His personal site offers a coffee address, but does not present it as a funding inbox.
Ask whether multiple users and applications can compute over reusable encrypted state without exposing the underlying data or retreating into isolated silos.
Map each privacy technology to its strongest property—TEE speed, MPC collusion resistance, FHE expressivity and ZK verifiability—then design a hybrid stack around the actual threat model.
Require users and developers to reason about hardware, manufacturer, operator and supply-chain trust through open standards, attestations and reproducible builds.
The user-facing benefit of privacy must outweigh integration cost; aim for private-by-default infrastructure with web2-like performance and web3-like verifiability.
Choose chain and infrastructure primitives around the application's specific user culture and experience rather than maximizing abstract decentralization metrics in isolation.
Crypto's autonomous software needs verifiable signing, policy and compute infrastructure that developers can integrate quickly without sacrificing modern product experience.
Privacy infrastructure should converge on a broadly usable abstraction with open standards, reproducible builds and end-to-end verifiability, while bounding rather than denying hardware risk.
Application-specific chains become compelling when deliberately assembled from modular primitives around a community's own user-experience values rather than generic chain ideology.
MPC plus delegated zero-knowledge proving can make private computation fast and composable enough for identity, AI, healthcare and modern onchain applications.
No podcast appearances are linked yet. Ask Kit to check for interviews that reveal how this investor thinks.
Quote coverage is incomplete. Kit can look for direct comments on markets, founders, and investment criteria.
Aadharsh invests across applied crypto and decentralized AI at Archetype. He previously worked with Galaxy Ventures and Flipside and brings experience in crypto research and data-driven diligence.
TEEs, MPC and FHE each have distinctive strengths and weaknesses; hybrid architectures can combine speed, collusion resistance and expressivity into usable private shared state.
The next privacy stack must let many users and applications compute over reusable shared data without revealing it, overcoming Privacy 1.0's tradeoff between secrecy and composability.
Blockchains can coordinate agents, verify offchain work and support private decentralized AI, but the space needs comparable benchmarks and privacy primitives that handle probabilistic computation.