AI Infrastructure

Neural
Intelligence
On-Chain

Merkavian builds cryptographically verified AI infrastructure. We make neural network training trustless through zero-knowledge proofs and decentralized federated learning.

ZK-Verified Gradient Proofs
4+ Active Miners
Base. Sepolia Deployed
54% Live Trading Win Rate

What we build

Two systems at the frontier of AI and decentralized infrastructure.

Research Protocol

PoLChain

A proof-of-learning blockchain protocol that uses zero-knowledge proofs to verify gradient updates from distributed miners — making AI model training cryptographically trustless.

ZK Proofs Federated Learning Base Sepolia EZKL
Explore protocol →
Live System

Autonomous Trading

A self-evolving algorithmic trading system with multi-timeframe signal analysis, reinforcement learning, and an evolutionary engine that continuously generates and validates new strategies.

Reinforcement Learning Signal Stack Evolution Engine Kraken
View system →

Core infrastructure

Zero-Knowledge Proofs

Cryptographic verification of neural network computations without revealing training data or model weights. Built on Halo2 circuits via EZKL.

Federated Learning

Distributed model training across independent miners using Federated Averaging. No raw data sharing required — privacy preserved by design.

On-Chain Verification

Smart contracts deployed on Base Sepolia handle task posting, gradient submission, ZK proof verification, and token reward distribution.

Evolutionary Strategies

50+ parallel strategy experiments with automated evaluation, promotion, and mutation — continuously searching for better trading approaches.

Multi-Signal Stack

MTF trend confirmation, VWAP bands, funding rate analysis, support/resistance detection, and pattern recognition — all validated against 5.4M candles.

Adaptive Sizing

Kelly Criterion position sizing with six dynamic multipliers including confidence, volatility, pair quality, and consecutive performance tracking.

Building at the frontier

Interested in our research or technology? We'd like to hear from you.

jackson@merkavian.com

PoLChain

A proof-of-learning protocol demonstrating that AI model training can be made cryptographically verifiable on a live blockchain — without a trusted aggregator.

The Problem

Leading AI companies face a fundamental economic crisis. In 2024, OpenAI spent $3B on model training against $3.7B in revenue. Anthropic spent $1.5B against $2.55B. Profitability under centralized training economics is nearly impossible.

Blockchain technology offers a different model — AI developers could issue tokens tied to model development, allowing funding to come directly from the market rather than burning capital. Miners earn tokens by contributing verified gradient updates. Token holders benefit as the model improves.

The critical unsolved problem: without a trusted aggregator, how do you verify that miners actually trained on real data and achieved the score they claimed?

The Solution

Zero-knowledge proofs make fraud cryptographically impossible. Each gradient submission is paired with a ZK proof that confirms training occurred correctly — without revealing any training data or model weights.

PoLChain demonstrates this system end-to-end on a live blockchain. Four automated miners compete each block, submitting gradients alongside Halo2 ZK proofs generated via EZKL. The best verified gradient wins $POL tokens.

polchain — block lifecycle
Block #1,204 opened · task posted to TaskManager.sol
Miners have 30s to submit gradient + ZK proof

Alpha submits: accuracy=94.2% · proof verified ✓
Beta submits: accuracy=93.8% · proof verified ✓
Gamma submits: accuracy=91.1% · proof verified ✓
Delta submits: accuracy=94.7% · proof verified ✓

Block finalized · Winner: Delta
FedAvg applied · global model updated · 100 $POL rewarded
New global accuracy: 95.1% (+0.4%)

Architecture

Five components work together in each block cycle:

  • Smart contract layer on Base Sepolia (POLToken, TaskManager, Verifier)
  • Four automated miners each training on independent MNIST data shards
  • Flask proving server generating Halo2 ZK proofs via EZKL
  • Quality-gated Federated Averaging — updates only applied if validation accuracy improves
  • Focus-digit system — miners oversample the model's current weakest digits each block

Technical Contribution

The central contribution is applying ZK proofs to neural network gradient verification. This requires converting floating-point ML operations into integer arithmetic over a finite field — what we call the gradient-gap problem. For simpler models like MNIST, the quantization error is small enough to be inconsequential. Solving gradient-gap at scale is the prerequisite for applying this to production models.

Autonomous
Trading

A self-evolving algorithmic trading system that uses multi-signal analysis, reinforcement learning, and an evolutionary engine to continuously improve its own strategies.

System Overview

The system runs continuously on a dedicated server, monitoring 15 cryptocurrency pairs across Kraken. It evaluates every potential trade through a seven-layer signal stack before execution, and runs 50 parallel experimental strategies simultaneously — killing underperformers and promoting winners.

The architecture is designed around one principle: no human intervention required. The bot learns from every trade, adapts its parameters automatically, and improves its own signal weights based on historical performance.

Signal Stack

Every trade consideration passes through seven independent signal layers before execution:

  • Multi-timeframe trend confirmation — 5m, 15m, and 1h alignment required
  • VWAP standard deviation bands — entries preferred at 2σ below fair value
  • Funding rate analysis — avoids overcrowded positions via Binance perp data
  • Candlestick pattern recognition — 7 patterns including hammer, engulfing, doji
  • Support/resistance proximity — +2.2% win rate improvement validated on 5.4M candles
  • Supertrend filter — 30% confidence penalty in confirmed bear regimes
  • Order book imbalance — Kraken depth analysis on high-confidence signals only

Evolutionary Engine

50 parallel sandbox experiments run at all times, each testing a mutation of the champion strategy. Every six hours the evaluator scores all experiments and kills underperformers. The experimenter then spawns new mutations weighted by lessons learned from historical training.

Historical training runs against 5.4M candles across 7 market regimes — bull, bear, sideways, high volatility, deep bear, crypto winter, and the last 30 days — to validate strategy robustness before promoting to champion.

oracle server — live cycle
Cycle #1,847 · 2026-04-21 15:04:22 UTC
Regime: NEUTRAL · 15 pairs · 50 experiments

BTC/USD RSI=38.4 · MTF=aligned · support=near
Confidence: Base=0.65 × MTF=1.20 × VWAP=1.15
→ Final: 0.71 · above threshold · entering

BUY BTC/USD @ $75,651 · size=$820 (8.2%)
Stop: $74,521 · Target: R1 $76,890 (dynamic)
ATR: $892 · Sizing: conf=1.18x adapt=1.0x

Reinforcement Learning Loop

The system learns from every trade through a closed feedback loop. Trade forensics record 15+ signal fields per closed trade. The self-tuner analyzes effectiveness after 20 trades and adjusts signal weights. The experimenter reads signal performance and boosts mutations that match what's working in live conditions.

Building trustless
AI infrastructure

Merkavian was founded on the belief that the next breakthrough in AI economics is decentralization. The current model — where a handful of companies bear the entire cost of model training — is fundamentally unsustainable.

We're building the infrastructure layer that makes distributed AI training cryptographically verifiable. Our research prototype, PoLChain, demonstrates that zero-knowledge proofs can solve the honesty problem in federated learning — making it possible for anyone to contribute compute and earn tokens without requiring a trusted central aggregator.

Alongside our research, we develop production systems at the frontier of applied AI — including an autonomous trading system with a self-evolving signal stack validated against 5.4 million historical candles across four years of market data.

Merkavian is headquartered in Berkeley, California, and founded by a student researcher at UC Berkeley working at the intersection of cryptography, machine learning, and decentralized systems.

Jackson Geiger
Founder & Researcher
UC Berkeley student researcher focused on the intersection of zero-knowledge cryptography, federated learning, and decentralized AI infrastructure. Author of the PoLChain proof-of-learning protocol.
Founded 2026
Location Berkeley, CA
Stage Pre-seed / R&D
Focus AI + Blockchain Infra
Email jackson@merkavian.com