Verified
This profile is verified using an email address.
ABHIJIT ROY
Bengaluru
Summary
Senior AI Solutions Engineer with hands-on experience helping enterprise customers deploy and optimize LLM workloads in PoC and production environments. Experienced in LLM inference and AI workload deployment. Skilled in translating customer requirements into practical AI deployment solutions, including solution architecture, model deployment, performance benchmarking, and technical enablement.
Overview
11
11
years of professional experience
Work History
AI Solution Engineer(SSME)
Red Hat
Bengaluru
10.2024 - Current
- Optimized AI deployments for customers by balancing workload requirements, performance, infrastructure utilization, and cost, enhancing overall efficiency and inference speed.
- Deploy and tune LLM inference workloads using vLLM and llm-d, adapting inference configurations to workload requirements and troubleshooting deployment and runtime issues.
- Benchmark LLM inference workloads using GuideLLM and evaluate key performance metrics to understand workload and deployment characteristics.
- Troubleshoot NVIDIA GPU issues on OpenShift, including NVIDIA GPU Operator configuration and related deployment issues, while supporting RHOAI upgrade and migration activities.
- Enable AI security and guardrail capabilities using NeMo Guardrails on RHOAI.
- Set up Models-as-a-Service (MaaS) on RHOAI to support customer AI application and inference requirements.
- Guide customers through the setup, configuration, upgrades, and migrations of Red Hat OpenShift AI (RHOAI) and Red Hat AI Inference Server (RHAIS) for their AI workloads.
- Delivered tailored RHOAI demonstrations aligned with customer requirements and use cases, facilitating technical discussions and proof of concepts.
- Developed technical solutions for enterprise customers, focusing on model selection, GPU planning, inference architecture, and AI workload deployment based on specific workload requirements and deployment constraints.
Software Maintenance Engineer
Red Hat
01.2018 - 01.2024
- Engineered LDAP/Identity solutions on RHEL, specializing in SSSD and Identity Management (IdM/IPA), enhancing enterprise authentication and directory integration.
- Resolved complex enterprise customer issues related to RHEL authentication, identity management, LDAP integrations, system performance, and patching.
- Collaborated with customers to pinpoint product issues and improvement opportunities, contributing to product enhancements based on real-world requirements.
Project Engineer
Wipro
01.2016 - 01.2018
- Managed enterprise Linux environments, addressing customer issues, troubleshooting systems, applying patches, analyzing performance, and automating operations.
- Coordinated Wipro project plans with cross-functional teams and client stakeholders.
- Tracked schedules, milestones, and dependencies using project management tools.
- Prepared status reports for Wipro leadership and delivery teams.
Education
Master of Computer Application -
Dibrugarh University
01-2015
Skills
- AI / LLM
- LLM Inference
- vLLM
- llm-d
- GuideLLM
- AI Model Optimization
- Model Deployment
- AI platform deployment
- Model Quantization
- Red Hat OpenShift AI
- Red Hat AI Inference Server
- Python
- RHEL
- SSSD
- Identity Management (IdM/IPA)
- LDAP
Personal Information
Title: AI Solutions Engineer / SSME
SELECTED TECHNICAL CONTRIBUTIONS
- IBM Tech 2026 — Selected Invitee — Selected as an IBM Tech 2026 invitee, recognizing technical impact and contributions to innovation and product development.
- PyCon India 2025 — Speaker — Green AI at Scale: Energy-Efficient LLM Serving using vLLM & LLM Compressor
- DevConf.IN 2025 — Technical Contributor — Showcased a customizable educational assistant built with Small Language Models (SLMs), exploring accessible AI-powered education.
- Red Hat — Technical Video Integrating RHEL with Active Directory using realmld and SSSD https://www.youtube.com/watch?v=CGt6oHZPToY
- Red Hat Developer — Technical Article - How PagedAttention Resolves Memory Waste of LLM Systems https://developers.redhat.com/articles/2025/07/24/how-pagedattention-resolves-memory-waste-llm-systems
- SSSD — Technical Article - Common issues in SSSD https://sssd.io/troubleshooting/errors.html
- VLLM — Open Source Contributor - Contributed native support for jina-embeddings-v5-text-small.
Timeline
AI Solution Engineer(SSME)
Red Hat
10.2024 - Current
Software Maintenance Engineer
Red Hat
01.2018 - 01.2024
Project Engineer
Wipro
01.2016 - 01.2018
Master of Computer Application -
Dibrugarh University