source: techcrunch ai: datadog veterans launch ai coding startup niteshift on a bet against big ai lock-in
level: business
niteshift, founded by early datadog engineers sajid mehmood and conor branagan, raised a $7 million seed round led by greylock's jerry chen. the startup enters the crowded ai coding market with a focus on infrastructure that separates the coding model from orchestration. it routes tasks between models like claude code and codex, plus open source options, based on project needs. the founders argue that companies will want a neutral vendor as model makers like openai and anthropic move into vertical software, potentially competing with their own customers.
the idea draws from datadog's early growth, when e-commerce companies avoided aws due to amazon's retail competition. mehmood sees a similar pattern now, with ai labs expanding into legal, healthcare, and finance. niteshift charges per-minute usage like a cloud provider, selling infrastructure rather than tokens. this model aims to let teams invest in developer tooling without locking into a single model or agent vendor.
the market already includes cursor, cognition, amazon bedrock, and openrouter, but niteshift bets on its team's experience scaling datadog. mehmood and branagan lived through the growing pains of large engineering orgs, which now face challenges with ai-generated code. they emphasize the need for infrastructure that can run, test, and verify software autonomously in production environments, built by people who have done it at scale.
why it matters: it offers a way for companies to use ai coding tools without depending on a single model provider, reducing risk as ai labs enter competing markets.
source: techcrunch ai: datadog veterans launch ai coding startup niteshift on a bet against big ai lock-in