StartupSan Francisco, CA · Seed

Crafting

Production-like environments with scoped access to real infrastructure for coding agents.

Crafting provides secure agent infrastructure running in the customer's cloud: sandboxes that mirror production, giving agents scoped access to database snapshots, internal services, caches and queues, with network isolation and credentials injected only when needed and never exposed to the model. Crafting for Agents, generally available since March 2026, adds Kubernetes interception, multi-service orchestration, hot-swapping of services and multi-region failover, complementing existing CI/CD. San Francisco-based; raised a $5.5M seed led by Mischief with WndrCo (March 2026). CEO Sumeet Vaidya.

Visit crafting.dev

Pros & cons

Each point cites its sources and says what kind of evidence backs it. A second reviewer tried to refute every claim before it was kept.

Pros

DocsAgents work against database snapshots, internal services, caches and message queues, with access scoped per agent by admins.[1]
DocsCredentials are injected at the network layer only when needed and are never exposed to the model, and the platform is SOC 2 Type II audited.[1]
DocsRuns inside your own cloud, multi-cloud and multi-region, with automated failover; Enterprise adds custom on-prem deployments.[1][2]
DocsHas a published customer case study: Faire calls it 'a step change in our developer experience'.[1]

Cons

DocsEvery plan, including Starter, is 'contact us for pricing': Starter is priced per seat and Business needs an annual contract.[1]
DocsBecause it runs in your own cloud, you provide and pay for the infrastructure and must govern the snapshot data agents see.[1]
DocsIt describes itself as 'agent-agnostic' but does not list supported agents, so test it with your harness.[1]
MetadataWe found almost no independent discussion of the agent product; the HN mentions are old lists of dev-environment tools.[1]

Best for: Platform teams that want coding agents to test against realistic internal services and data without being handed raw credentials.
Watch out for: Plan for governing snapshot data and for infrastructure costs in your own cloud.

Assessed Oct 29, 2026 · adversarially reviewedSuggest a correction

People

Sumeet Vaidya

Co-founder & CEO

Harness engineering

Yisui Hu

Co-founder & CTO

Harness engineering

San Francisco, CA · Est. 2023

Featured

Cognition

Devin, the AI software engineer, and the Windsurf IDE.

Cognition builds Devin, an autonomous AI software engineer that plans and executes engineering tasks end-to-end, and acquired the Windsurf agentic IDE in 2025. One of the defining startups of the coding-agent wave.

~$10.2B valuation (Sep 2025)

3 pros · 4 cons

Prague, Czechia · Est. 2023

E2B

Open-source sandboxes for running AI-generated code.

E2B gives agents secure, isolated cloud micro-VMs to execute code and commands — the execution runtime layer of the harness, used by many coding and data-analysis agents.

4 pros · 3 cons

Est. 2023

Daytona

Elastic sandboxes for running AI-generated code.

Daytona provides secure, OCI-compatible sandbox 'computers' that spin up in under 90ms — with snapshots for persistent agent state. In 2026 it moved core development to a private codebase while continuing to run as a commercial platform.