Build it right the first time
Greenfield systems in Haskell and Rust, designed so whole classes of bug cannot be expressed.
Learn moreWe build the systems
where bugs cost real money
High-assurance software engineering in Haskell and Rust,
backed by formal methods when correctness matters most.
For fintech, blockchain, and deep tech.
Across Industry & Academia
- 10
- years shipping production FP code since 2015
- $25B+
- combined client valuation
- 80
- engineers
- $600M+
- raised by our clients
- 8+
- blockchain networks, protocol level
Custom Software Development
What teams come to us for
Prove it's correct
Formal methods and machine-checked proofs, for the parts of a system where nearly right is not good enough.
Learn more
Rescue what's already built
Legacy Haskell nobody can maintain, performance bottlenecks, infrastructure that doesn't scale. We take over and stabilize.
Learn more
Add capacity you can't hire
Functional programming and protocol engineers, embedded in your team, on the problems you cannot staff for.
Learn moreWhere we go deeper than anyone
Haskell
We sponsor a full-time team of GHC contributors — we help build the compiler itself.
Rust
Our Rust flies on drones, powers biotech labs, and moves millions through DeFi.
Formal verification
Mathematical proofs of correctness for smart contracts and critical systems.
Blockchain
- Canton
- Solana
- Cardano
- Ethereum
- Midnight
- TON
- Tezos
- DFinity
- Bitcoin
AI / ML
From RAG systems to distributed GPU computation protocols. Backed by our in-house science lab.
Developer tooling
We wrote nixfmt and deploy-rs — the official Nix formatter and deploy tool, running in CI across thousands of repositories.
Industries
Fintech & Banking
Payment systems, trading platforms, and Canton/DAML smart contracts for institutions.
Learn more
Web3 & DeFi
Protocol-level engineering across 8+ networks: DEXs, bridges, wallets, audits.
Learn more
Deep Tech & Hardware
Rust for devices that can't fail: drones, lab instruments, embedded systems.
Learn more
Biotech
Software for research hardware and data pipelines, from spectroscopy to drug screening.
Learn more
Dev Tools
Compilers, build systems, and developer infrastructure — including GHC itself.
Learn moreSelected work
The Open Network
Developed an asynchronous payment channel and a multisignature wallet.
First prize in Telegram’s TON competition.
Cardano SL
Implemented the core protocol, network infrastructure, and launch tooling alongside IOHK.
From a research paper to a live blockchain.
Edna
Built an open-source MVP for biotech researchers to analyse, store, and reuse experiment data.
Experiment analysis and history in one workspace.
What our clients say
30+ verified reviews on Clutch
- Rust
- TypeScript
- Scientific hardware
Serokell has deep technical expertise in writing quality software, by using strongly typed languages that help you avoid certain classes of bugs and poor engineering practices from the start. […] These guys are smart and at the top of their game, and furthermore care about their craft.
Chief Software OfficerMRR spectroscopy instruments company 5.0 on Clutch · Nov 2023- Solana
- DeFi
We hired Serokell to access their expertise and to co-develop a Solana smart contract with our engineers, while also bringing Solana best practices into our team.
Product Owner1inch 5.0 · Feb 2026- TON
- Cross-chain bridge
A project manager oversaw the process, while the CTO supervised and reviewed every step. All the agreements were kept. The deadlines were met.
The solution has been launched and has successfully passed a security audit.
CPOSymbiosis Finance 5.0 · Mar 2025How we work
Technical from day one
Step 1Our first conversation includes people who can actually evaluate the problem.
Project management included
Step 2Every engagement has a bill. You don't pay a separate management markup.
Close communication
Step 3Shared Slack, weekly calls, clear ownership and status.
A team, not a talent marketplace
Step 4Typically 2–4 engineers working under Serokell technical supervision.
Latest News
& Posts
deploy-rs in 2026: reviving our Nix deployment tool
deploy-rs kept deploying our infrastructure while its maintenance backlog grew. This year, we returned to that backlog and tagged version 1.0.0, the first versioned release of the Nix deployment tool we introduced in 2020. Here’s what changed, how we use it to deploy systems and individual services, and what we plan to tackle next.
What Anthropic's AI Proofs Mean for Formal Verification
Some of the bugs Claude found in open-source libraries had survived trillions of fuzzing runs. It found them while trying to prove that the code could never crash on inputs within a specified size limit. Anthropic and Theorem used this approach to produce machine-checked proofs for 239 fuzz-test entry points across 123 libraries. The interesting question is what those proofs buy you. A proof can rule out a crash within the limits of its specification and machine model. Deciding whether those limits make sense for the software you actually run is a separate engineering job. We look at what Claude proved, what it took to get there, and where the work goes when proofs get cheaper.
Serokell to Take On Cardano Consensus Development
Serokell will take on the development and maintenance of Cardano’s consensus implementation as part of Input Output’s ongoing effort to decentralize the network’s core engineering..
Talk to us
Tell us what you're building. We'll tell you honestly whether we're the right team for it — on a 30-minute technical call.
Contact us