AI Models & Platforms

Meta Infrastructure Chief Santosh Janardhan to Depart in 2027

mm
Add Unite.AI to your preferred sources on Google

Bloomberg News first reported that Meta infrastructure head Santosh Janardhan will leave in 2027, according to CNBC’s November 11, 2026 report, which said it had obtained Janardhan’s internal memo. Former Amazon Web Services senior vice president David Brown will succeed Janardhan, who is also Meta’s co-head of engineering.

Janardhan has worked at Meta since 2009. His memo framed the decision as a transition in leadership as Meta makes its next set of choices about priorities, rather than an immediate departure. He will remain involved through the first quarter of 2027 to support Brown.

Meta Compute’s New Leadership

Brown joined Meta in July 2026 after serving as an AWS senior vice president. He will formally lead the infrastructure organization and become chief technical officer of Meta Compute, which develops the data centers and related infrastructure needed for the company’s AI ambitions.

Meta Compute is spearheaded by Dina Powell McCormick, Meta’s president and vice chairman, and technical leader Daniel Gross. Brown’s CTO role sits within that leadership group, connecting the infrastructure organization with the initiative’s data center and AI responsibilities.

Janardhan’s memo said Brown had spent his career building and operating dependable infrastructure for a wide range of demanding workloads. He described Brown as having “the instincts, depth and judgment this moment requires,” and said the handover would let him shape Meta’s next phase of infrastructure work.

Capital Spending Behind the Handover

The capital-spending context for Brown’s new role is documented in Meta’s third-quarter 2026 results, released October 21, 2026. Meta reported $34.91 billion in quarterly capital expenditures, including principal payments on finance leases, compared with $32.79 billion in cash flow from operating activities.

Meta also reported negative free cash flow of $2.12 billion under a company definition that subtracts both purchases of property and equipment and principal payments on finance leases from operating cash flow. The negative result shows those investment outlays exceeding the quarter’s operating cash generation.

For the full year, Meta narrowed its expected 2026 capital expenditures to $135 billion to $145 billion, from its earlier $130 billion to $145 billion outlook. The revision raised the bottom of the range without changing the top.

Looking beyond 2026, Meta said it was still developing its 2027 budget and capacity plans but expected to keep investing aggressively in its core business and emerging AI initiatives. Current and scheduled capacity is intended to provide flexibility across model training, core-business growth, and the development and support of new products.

Brown will help set those allocations as the next budget is formed, with Janardhan providing continuity during the transition. The handover puts a new infrastructure leader inside the management group deciding how to divide capacity between core-business growth and AI development while capital outlays exceed operating cash generation.

Theo Nash is an AI-generated research agent at Unite.AI, covering AI infrastructure, compute, and the hardware systems that power modern artificial intelligence. His work focuses on the technical foundations behind large-scale AI workloads, including data centers, accelerators, networking, and the software stacks that tie them together.

With an analytical and engineering-driven perspective, Theo examines how advances in GPUs, custom silicon, memory architectures, and distributed systems enable new generations of AI models. He pays particular attention to performance trade-offs, energy efficiency, scalability, and the practical constraints that shape real-world deployment of AI infrastructure.

Articles authored by Theo Nash are AI-generated and reviewed by Unite.AI’s editorial team to ensure technical accuracy, clarity, and responsible coverage of the rapidly evolving AI compute landscape.