Bitcoin Builders ★ 4.0

About us
Bitcoin Builders is a leading company in Blockchain and Artificial Intelligence (AI). We create innovative solutions that combine Blockchain and AI to improve processes and user experiences. We are looking for an talented AI Engineer for our products.

Role Summary

You will design, build, and operate a fully distributed, peer‐to‐peer (P2P) infrastructure in which every node can host a large‐language‐model (LLM) and expose real‐time inference APIs to users. Your mission is to make the network self‐orchestrating, fault‐tolerant, secure, and cost‐efficient while delivering low‐latency, high‐throughput model serving at scale.

Key Responsibilities

Architecture & Design

• Define the overall P2P topology, discovery, and routing protocols.
• Partition/model‐sharding strategies (tensor, expert & pipeline) to fit heterogeneous devices.
• Establish secure, trust‐minimized mechanisms for node onboarding, reputation, and payment/credit settlement.

Model Serving & Optimization

• Package and serve multiple LLM checkpoints (open‐source and custom) across CPU, GPU, and accelerator back‐ends.
• Implement quantization, speculative decoding, KV‐cache offloading, and other inference‐speed techniques.
• Monitor latency, token‐throughput, and memory footprint; drive continuous optimisation.

Distributed Systems Engineering

• Build gossip‐based or DHT‐based neighbor discovery; maintain consistency of model versions across the swarm.
• Design streaming and chunked transport layers to support long-context or streaming‐generation workloads.
• Handle churn gracefully (node joins/leaves, flaky devices, NAT traversal, intermittent connectivity).

- 2 - 3+ years building production distributed systems or networking software.
- Strong Golang or Rust backend engineering skills (plus working knowledge of Python for ML workflows).
- Proven experience with P2P overlays (e.g., libp2p, BitTorrent, Kademlia, GossipSub) or decentralized‐compute frameworks (Golem, Folding@home, etc.).
- Hands‐on with LLM inference stacks (vLLM, TGI, TensorRT‐LLM, HuggingFace Text Generation Inference, or similar).
- Familiarity with GPU/accelerator performance tuning, CUDA, Triton, or ROCm.

Nice‐to‐Have

• Experience contributing to prominent open‐source P2P or blockchain projects.
• Knowledge of token‐based economic incentive models, micro‐payments, or proof‐of‐stake mechanisms.
• Familiarity with federated learning, model‐weight distillation, or on‐device fine‐tuning.
• Prior work with RAG pipelines, or vector database integration.
• Publications or talks on large‐scale inference or distributed AI.

• Competitive salary package
• Flexible working time and open, dynamic environment with stunning colleagues
• Healthy lunches and free flow of fresh milk, coffee, tea, and more while working in our fabulous office
• Sick leave, holidays, and at least 12 paid days off annually.
• Full medical and dental coverage for you and 50% for your family
• MacBook Pro, monitor, smart desk & chais and any accessories provided to do your best work
• Free access to a wide variety of online courses

• Competitive salary package
• Flexible working time and open, dynamic environment with stunning colleagues
• Healthy lunches and free flow of fresh milk, coffee, tea, and more while working in our fabulous office
• Sick leave, holidays, and at least 12 paid days off annually.
• Full medical and dental coverage for you and 50% for your family
• MacBook Pro, monitor, smart desk & chais and any accessories provided to do your best work
• Free access to a wide variety of online courses