Datadog Veterans Launch Niteshift: A New Dawn for AI Coding Startups

TL;DR

• Niteshift, founded by former Datadog engineers Sajid Mehmood and Conor Branagan, has raised a $7 million seed round led by Greylock.
• The startup is building infrastructure for AI coding agents so companies can switch models without getting locked into one provider.
• Its pitch lands in a fast-moving market where AI labs are moving up the stack, raising demand for more control, portability, and autonomy.

Datadog Veterans Return With a New Bet on AI Infrastructure

Two former early Datadog engineers are back with a startup aimed at one of the biggest anxieties in enterprise AI: dependence on a single model vendor. Sajid Mehmood and Conor Branagan have launched Niteshift, an AI coding agent startup that just closed a $7 million seed round led by Greylock’s Jerry Chen. The round also drew notable angels including Reid Hoffman and Datadog co-founders Olivier Pomel and Alexis Lê-Quôc.

The company’s central argument is straightforward: as frontier model providers become more capable and expand into application layers, startups and enterprises risk being squeezed out or locked into ecosystems they do not control. Niteshift is positioning itself as the infrastructure layer that lets customers build and run coding agents without tying their operations to one model maker.

Why Niteshift Thinks Lock-In Is the Real Problem

Niteshift is not trying to compete by launching yet another foundation model. Instead, it is selling the “plumbing” that sits underneath AI coding workflows, with the goal of making it easier to route tasks across multiple models depending on cost, performance, or project needs.

That pitch speaks directly to a broader industry concern: if a company’s coding agents, developer tooling, and workflow logic are all embedded in one provider’s stack, switching later can become expensive and operationally risky. Niteshift says its platform is meant to preserve customer autonomy by allowing businesses to invest in their own tooling while avoiding dependence on a single vendor.

A Cloud-Like Model for Coding Agents

According to the company’s description, Niteshift intends to charge in a way that resembles cloud infrastructure pricing, using per-minute usage rates rather than a model-centered subscription tied to one AI vendor. That approach suggests the company wants to be viewed less as a chatbot app and more as a utility layer for production-grade coding agents.

This matters because AI coding is increasingly moving from demos and prototypes into real software development workflows. In that environment, procurement teams and engineering leaders tend to care less about novelty and more about reliability, portability, and predictable costs. Niteshift’s pricing and product strategy appear designed to align with those priorities.

The Market Context: AI Labs Are Moving Up the Stack

The timing of Niteshift’s launch is notable. The biggest AI model companies are no longer just supplying raw models; they are increasingly building consumer and enterprise products on top of them. That creates an opportunity for infrastructure startups that can offer a neutral layer between model makers and customers.

Greylock’s Jerry Chen framed the bet in those terms, saying that as frontier labs move up the stack, there is room to offer an alternate path by unbundling agents from the infrastructure they run on. In that view, Niteshift is not merely another AI startup; it is a response to the strategic realignment of the AI market itself.

Why the Founders Matter

Mehmood and Branagan bring credibility from their time helping Datadog grow from its early days into a multi-billion-dollar company. That background likely helps Niteshift with both technical execution and customer trust, especially in a category where buyers are wary of betting critical workflows on unproven tools.

The presence of Datadog’s founders among the angels is also symbolically important. Datadog built its reputation on observability and infrastructure for modern software systems; backing Niteshift suggests confidence in another infrastructure-first thesis, this time for AI coding agents.

What Comes Next

Niteshift still faces a crowded and fast-changing field. AI coding is one of the most competitive categories in startup land, and incumbents as well as model providers are moving quickly to own more of the developer workflow.

Still, the company’s focus on vendor neutrality may be its sharpest differentiator. If enterprises continue to worry about being trapped inside the product strategy of a single AI lab, infrastructure that enables portability could become increasingly valuable. Niteshift is betting that autonomy will matter as much as capability in the next phase of AI software development.

AndroGuider Team

Articles written by the AndroGuider team. We try to make them thorough and informational while being easy to read.