# Mrityunjay Kumar

> Senior ML engineer building production ML systems, agent infrastructure, evaluation pipelines, and reliable, high-throughput distributed serving systems.

## Professional profile

- Role: Senior ML Engineer at Microsoft
- Focus: ML Systems & Agent Infrastructure
- Experience: 12+ years across ML systems, distributed services, storage, and data infrastructure

## Engineering focus

- Distributed infrastructure for reliable LLM agents
- Inference engines, rigorous model evaluation, and RL systems at scale
- Zero-to-one architecture, implementation, observability, and production delivery

## Current engineering work

- Scaled the agent's distributed execution and serving infrastructure to thousands of concurrent containers under strict latency SLOs, with load balancing, autoscaling, and automated fault detection and recovery.
- Built PowerPoint's first end-to-end fine-tuned model workflow, including synthetic-data generation, batched data preparation, secure deployment, inference, and integration with PowerPoint workflows.
- Designed extensible slide-generation and evaluation systems with configurable models and datasets, end-to-end telemetry, and parallel execution that reduced evaluation time by 4.6x.
- Designed and operated PowerPoint cloud services across commercial and government environments, adding automated validation, observability, and incident-response tooling.

## Core expertise

- Production ML systems
- LLM agent infrastructure
- Model evaluation
- Fine-tuning pipelines
- PyTorch
- Hugging Face Transformers
- Synthetic data generation
- MLOps
- Distributed systems
- Inference serving systems
- Load balancing & request routing
- Autoscaling & orchestration
- Latency & throughput optimization
- Fault detection & recovery
- Distributed RL / post-training infrastructure
- Consensus protocols (Paxos, Raft)
- Transactional databases
- Crash consistency
- Cloud-native microservices
- High-throughput pipeline design
- Azure
- Azure Machine Learning
- Azure Container Instances
- AWS
- Docker
- Redis
- Kafka
- gRPC
- OpenTelemetry
- Prometheus
- Grafana
- Python
- C#
- C++
- C
- Java
- Go

## Research and intellectual property

- [Rolis: a software approach to efficiently replicating multi-core transactions](https://dl.acm.org/doi/10.1145/3492321.3519576): EuroSys 2022 · systems implementation & evaluation
- [Verification of metadata consistency across snapshot COW B+ tree logical maps](https://patents.google.com/patent/US11573860B1): US11573860B1
- [Learning to fingerprint the latent structure in question articulation](https://ieeexplore.ieee.org/document/8614044): R. Guntur, M. Kumar · IEEE ICMLA 2018 · pp. 73–80

## Canonical sources

- [Professional profile](https://mrityunjaykumar911.github.io/)
- [Detailed profile: RL & evaluation infrastructure](https://mrityunjaykumar911.github.io/latest.html)
- [Source repository](https://github.com/mrityunjaykumar911/mrityunjaykumar911.github.io)

## Privacy

This context intentionally excludes personal contact details and private employment records.