Headline

MLOps @ Vimaan

About

Experience in building Production ML Systems, Natural Language Processing, Computer Vision, Python real-time software solutions. Open to any kind of MLOps, Machine Learning / Data Solutions in the respective fields. topmate.io/raj_srujan_jalem/?utm_source=topmate&utm_medium=popup&utm_campaign=Page_Ready

Location

Bengaluru

AI stack

claude code

codex

deepseek

hermes agent

ollama

opencode

Work experience

Manager, MLOps at Vimaan, Aug 2025 - Present · 1 y 3 mo, Bengaluru, – Architected and led (team of 9) the end-to-end ML inference platform across edge capture, streaming, and GPU serving; cut inference latency 70% via pipeline profiling and parallelized pre/post-processing. – Built a distributed, on-premise edge AI pipeline on ROS for real-time multi-camera ingestion, decoding, and inference, processing 1–2 TB of video per day across edge compute nodes. – Cut edge bandwidth with event-driven frame sampling (detection-triggered 4K capture); tuned ROS 1/ROS 2 DDS QoS and transport via Wireshark packet analysis to eliminate packet loss. – Scaled GPU model serving on Triton Inference Server and Ray Serve: TensorRT engines, FP16/INT8 quantization, dynamic batching, and concurrent model instances, raising GPU utilization and cutting compute cost within real-time latency SLAs. – Built a production observability stack (OpenSearch, Prometheus, Grafana) with centralized logging, metrics, alerting, and automated root-cause analysis across distributed microservices. – Standardized dev/test/prod deployment patterns with Helm, Terraform, and CI/CD, enabling versioned model rollouts and rollback across hybrid edge-cloud infrastructure. – Built GenAI platform capabilities including RAG pipelines on vector search, MCP-based tool integration, and LLM evaluation workflows for internal agentic applications.

Staff Engineer - MLOps at Vimaan, Mar 2025 - Aug 2025 · 6 mos

Senior MLOps Engineer at Vimaan, Jan 2025 - Apr 2025 · 4 mos

Backend Engineer - ML at Alectio, Jan 2022 - Aug 2022 · 7 mo, Remote, – Built the Alectio Model Library, a reusable model integration SDK used across customer teams to standardize model onboarding, automated training, and active-learning workflows. – Built automated labeling pipelines that reduced customer data annotation costs by ~40%. – Built core backend microservices for Alectiolite and mentored junior developers on ML infrastructure best practices.

AI Data Engineer at Fasal - Grow More, Grow Better™, Mar 2021 - Jan 2022 · 10 mo, Bengaluru, Karnataka, India, - Set up automated forecasting and continuous training workflows on Kubernetes using Kubeflow and the MLflow model registry for model versioning and promotion. – Migrated monolithic pipelines to a microservice architecture and implemented a Feature Store, reducing API latency by 70%. – Streamlined CI/CD pipelines and established robust Docker container log analysis, speeding up deployments 3x and shrinking container image sizes by 80%.

Data Science Engineer at Utopia Global, Inc., Jan 2019 - Mar 2021 · 2 y 2 mo, Greater Bengaluru Area, – Trained custom U-Net models for super-resolution image enhancement, doubling the output resolution quality. – Refactored production vision pipelines to increase processing throughput by 50%. – Developed custom NLP and NER models for a resume parser deployed on AWS EC2, and mentored a team of 6 junior data scientists.

Organizer at PyData Raipur, Aug 2017 - Nov 2019 · 2 yrs 4 mos, Greater Raipur Area

Education

Bachelor’s Degree, Computer Science, IIIT-Naya Raipur

High School, Junior High/Intermediate/Middle School Education and Teaching, Delhi Public School, visakhapatnam

Public links

github: https://github.com/rajsrujan77