W
Wand builds a platform for deploying AI agents alongside human workers in enterprise settings. The tech stack reveals a data-intensive, ML-forward architecture: LLM orchestration (LangChain, LangGraph, LiteLLM, OpenAI), vector storage (Pinecone, Weaviate, Qdrant), streaming (Kafka, RabbitMQ), and distributed compute (Kubernetes, Temporal, Databricks). Current pain points center on production scaling and agent reliability—hiring skews heavily engineering (19 of 24 open roles, mostly senior/staff level), signaling active buildout of infrastructure and core platform capabilities rather than go-to-market expansion.
Tech Stack 34 technologies
Core StackAWS Terraform Kubernetes React JavaScript Python FastAPI Temporal Elasticsearch MongoDB PostgreSQL Redis ClickHouse Snowflake OpenAI LangChain RabbitMQ Kafka Docker BigQuery Databricks Pinecone Weaviate GraphQL Miro Azure LangGraph LiteLLM LlamaIndex Qdrant+4 more
What Wand AI Is Building
◆Challenges
- Scaling production infrastructure
- Rebuilding infrastructure and processes
- Adopting next generation ai systems
- Improving uptime and reliability
- Adoption gaps
- High-performance distributed systems
- Ensuring agent reliability
- Scalable ml infrastructure for autonomous ai agents
- Improving reliability at scale
- Retrieval pipeline latency
▲Active Projects
- Productionise ml workloads
- Control panel platform
- Scalable backend systems
- Design and implement workflows, agents, and integrations on wand for live customer scenarios
- Strengthen kubernetes-based architecture
- Optimise ci/cd pipeline
- Agent skill creation and lifecycle management
- Embedding models into applications
- Agent evaluation and feedback loops
- Scalable ml platform for autonomous ai agents
Hiring Activity
25 roles · 25 in 30d
Department
Seniority
Notable leadership hires: Head of Engineering
Company intelligence