Type to search posts and projects ↑↓ to navigate

Subhadip Mitra · Google Cloud

Data platforms. AI systems. The infrastructure between them.

Head of Data & Analytics and Site Lead at Google Cloud, leading Data & AI innovation and transformation across Southeast Asia. I publish research on multi-agent systems, inference optimization, and AI safety, and write the practitioner's notes behind it, from inside real production systems.

21 publications · 10 packages LatestICLR 2026 Workshop Spark-LLM-Eval

01 Now

Building

ICLR 2026 workshop paper on LLM safety, and compute primitives for orbital environments.

Exploring

New ventures at the frontier; conversations with research labs and founders.

Writing

Fused MoE kernels, circuit tracing in production, and bets on model honesty.

What I'm up to →

02 Latest

Your OpenAPI Spec Is Not an Agent Interface: Compiling Governed MCP Tool Surfaces

§ MCP · Aug 2026

api-mcp-compiler turns OpenAPI, Swagger, and WSDL specifications into MCP tool surfaces with field-level provenance, human approval gates, and compiled-in refusals. Four pre-registered evaluations could not resolve wh...

Read the essay →

03 Selected writing

DEEP-LEARNINGJul 2026 Beating FP16 with 4-bit Weights: A Portable W4A16 GEMM in TritonI wrote a 4-bit weight-only GEMM in pure Triton. The fast W4A16 kernels are all CUDA, so this... AI-SAFETYJul 2026 The Activation-Cone Blind Spot, or Why Your Jailbreak Defense Can't See PrefillingPrompt-time activation defenses stop GCG and AutoDAN cold, then fail half the time against pr... INTERPRETABILITYJul 2026 What Runtime Interpretability Actually Costs, Part 1: The Case for Measuring ItEveryone assumes activation probes are too expensive to run in production. I ran the numbers ... DEEPSEEKJun 2026 DeepSeek DSpark: Speculation Is a Scheduling ProblemWhat DeepSeek's DSpark and DeepSpec release actually changes for LLM inference: suffix decay,... AGENTSJun 2026 Loop Engineering: Convergence Is Not CorrectnessA loop always converges. That it reached a stable 'done' state tells you nothing about whethe...
All essays in the archive →

04 Research focus

/ 01

Multi-agent systems

Coordination, debate, and verification for AI that runs unattended.

 / 02

Inference optimization

Fused MoE dispatch, speculative decoding, custom Triton kernels.

 / 03

AI safety & interpretability

Circuit tracing, activation steering, sandbagging detection.

 / 04

Distributed systems

Formal synthesis and correct-by-construction architectures.

05 Instruments

INFERENCEInteractive Memory-bandwidth rooflineMemory-bound or compute-bound, and how KV cache and kernel efficiency move the wall. ADVERSARIALInteractive Quality-diversity archiveWhy MAP-Elites maps diverse vulnerabilities that a single objective misses.
All instruments →

06 Selected publications

2026
Closing the Activation-Cone Blind Spot: Response-Time Probing and Unified Defense
arXiv
PDF →
2026
Cross-Generational Transfer of Adversarial Attacks Reveals Non-Monotonic Safety Alignment in LLMs
arXiv
PDF →
2026
Cross-Platform Fused MoE Dispatch in Triton: Portable Expert Routing Without CUDA
arXiv
PDF →
2026
Spark-LLM-Eval: A Distributed Framework for Statistically Rigorous Large Language Model Evaluation
arXiv
PDF →
2026
Quality-Diversity Evolution for Discovering Diverse Vulnerabilities in LLM Safety
ICLR 2026 Workshop AIWILD
PDF →
All publications →
Subhadip Mitra