Senior DS/ML developer
About Streamflow
Streamflow is the leading token distribution infrastructure on Solana, trusted by over 40,000 projects and managing more than $1.4 billion in total value locked (TVL). We help crypto teams automate vesting schedules, airdrops, staking pools, and payroll — replacing spreadsheets and manual transfers with audited smart contracts.
Our mission is simple: make token distribution seamless, transparent, and programmable for every Web3 team.
We are now scaling our data team to support rapid growth across Solana, Aptos, and Sui ecosystems. We are profitable, fully remote, and growing fast.
The Role
We are looking for a Senior Data Scientist & ML Developer to join our analytics team. This is a hybrid role — part data science, part lightweight ML engineering — focused on unlocking insights from on-chain data and building predictive models that help our users and internal teams make better decisions.
You will work with real blockchain data from over 1.3 million users and 40,000 projects. You will build models that predict user churn, detect anomalous vesting patterns, optimize staking reward distributions, and identify fraudulent airdrop claims.
This is not a pure research role. You will write production-ready code, deploy lightweight models, and collaborate closely with product managers and engineers.
What You Will Do
Predictive Modeling
Build models to forecast user retention, vesting claim behavior, and staking participation rates.
Develop churn prediction for project accounts — helping us understand which teams are likely to stop using Streamflow.
Anomaly Detection
Create systems to detect suspicious on-chain activity: sybil attacks on airdrops, wash trading in staking pools, or unusual token movement patterns.
Work with the compliance team to flag high-risk behavior without compromising user privacy.
On-Chain Data Science
Analyze massive datasets from blockchain using SQL and Python.
Identify trends — which vesting schedules perform best? What drives higher claim rates?
ML Engineering (Lightweight)
Deploy and maintain small-scale ML models in production using batch inference (no real-time requirements).
Collaborate with data engineers to ensure clean, reliable data pipelines.
Write maintainable, documented code in Python using libraries like scikit-learn, XGBoost, or LightGBM.
Strategic Input
Present findings to leadership and product teams.
Help define key metrics and success criteria for data science initiatives.
Requirements
Experience – 5+ years in data science, machine learning, or analytics engineering roles.
Python – Strong proficiency with pandas, numpy, scikit-learn, and at least one ML framework (XGBoost, LightGBM, or similar). You do not need deep learning (PyTorch/TensorFlow).
SQL – Expert-level SQL. You will query large on-chain datasets across multiple tables.
Blockchain Knowledge – You understand how blockchains work: transactions, blocks, wallets, smart contracts, token standards (SPL on Solana). You know the difference between a wallet address and a token account.
Production Mindset – You can write code that runs reliably in production. You understand batch inference, feature stores, and model monitoring.
English – Fluent written and spoken.
What Makes This Role Unique
Real on-chain data – Not simulated, not anonymized. You will work with billions of real transactions from millions of real users.
High-impact problems – Your models will directly help detect fraud, improve user retention, and optimize millions of dollars in token distributions.
Autonomy – Small team, no bureaucracy. You will own the entire data science lifecycle — from ideation to deployment to monitoring.
Crypto-native culture – We speak in TVL, APY, vesting cliffs, and lockup periods. You will learn tokenomics from the inside.
Working at Streamflow
100% remote – Work from anywhere in the world. No office mandates.
Asynchronous first – We write things down. We minimize meetings. Deep work is sacred.
Time zone overlap – We ask for approximately 4–6 hours of overlap with CET (Central European Time). The team is distributed globally, but we coordinate intentionally.
Startup culture – Fast decisions, high ownership, no micromanagement. Everyone ships.
Physical hub – Optional office space in Belgrade, Serbia if you ever want a desk and in-person collaboration.
Benefits
Fully remote – No return-to-office. Ever.
Flexible schedule – Outcomes matter, not hours logged.
Paid time off – 20+ days per year + local public holidays
Hardware stipend – One-time budget for your home office setup
Professional development – Budget for conferences, courses, or certifications
How to Apply
Your resume (PDF or LinkedIn link)
A brief note about your experience
