Prateek Verma profile photo 1 on Tal

Prateek Verma

AI Platform Architect | MLOps & LLMOps | Production GenAI & Agentic Systems | AWS/Azure/GCP | 35% Cost Optimization

Bengaluru

7+ years of work experience

Deloitte
Jul 202511 mosPresent

Bengaluru

Capgemini
Jan 20223 yrs 6 mosJun 2025

Pune, Maharashtra, India

Firstsource
May 20201 yr 7 mosNov 2021

Bengaluru, Karnataka, India

FUGO Services
Mar 20191 yr 3 mosMay 2020

Bengaluru, Karnataka, India

Innovsource
Oct 20186 mosMar 2019

Bengaluru, Karnataka, India

Prateek’s AI Stack

hugging face
hugging face
langchain
langchain
langgraph
langgraph
ollama
ollama

Join Prateek and 10k+ other professionals on

Tal Boss

Tal is the app where bosses meet engineers & builders directly, without middlemen

Download Tal Boss

Prateek Verma on Tal

Headline

AI Platform Architect | MLOps & LLMOps | Production GenAI & Agentic Systems | AWS/Azure/GCP | 35% Cost Optimization

About

🧠 AI Platform Architect | MLOps & LLMOps | Production GenAI & Agentic Systems I design and operate production AI platforms. Over 7+ years, I’ve worked across enterprise environments (Deloitte, Capgemini, Philips, Hitachi, Unilever), focusing on one thing: taking AI systems from experimentation to reliable, scalable production platforms. My work sits at the intersection of MLOps, LLMOps, and AI platform architecture — owning systems end-to-end: from orchestration design to deployment, governance, and continuous optimization. What I own in production AI platform architecture — multi-agent systems, orchestration layers, MCP integrations, and human-in-the-loop workflows RAG systems — retrieval design, grounding strategies, evaluation (RAGAS), and continuous quality improvement Model lifecycle governance — experiment tracking, versioning, deployment strategies, retraining, and drift monitoring Inference cost engineering — token optimization, context compression, and cost-aware architecture (35% cost reduction delivered) AI platform CI/CD — automated pipelines, release governance, and environment-aware deployments Production impact ⚡ 40% faster deployment cycles through pipeline architecture decisions 💰 35% reduction in inference cost through token and payload optimization 🧠 Continuous LLM quality improvement via structured evaluation pipelines 🔍 Full observability across ML + LLM systems (monitoring, drift, feedback loops) Platforms I work across MLOps: MLflow · Airflow · SageMaker · Kubeflow · Feast · Evidently AI LLMOps: LangChain · LangGraph · RAG pipelines · evaluation frameworks Cloud: AWS · Azure · GCP Infra: Kubernetes · Docker · Terraform I focus on production systems, not prototypes — where reliability, cost, governance, and scale matter as much as model performance. 🤝 Open to conversations around AI platforms, MLOps, LLMOps, agentic systems, and production GenAI architecture

Location

Bengaluru

AI stack

hugging face

langchain

langgraph

ollama

Work experience

Deputy Manager (Senior Consultant) at Deloitte, Jul 2025 - Present · 11 mos, Bengaluru

Professional I (Associate Consultant) at Capgemini, Jan 2022 - Jun 2025 · 3 yrs 6 mos, Pune, Maharashtra, India

Associate at Firstsource, May 2020 - Nov 2021 · 1 yr 7 mos, Bengaluru, Karnataka, India

Process Analyst at FUGO Services, Mar 2019 - May 2020 · 1 yr 3 mos, Bengaluru, Karnataka, India

Associate at Innovsource, Oct 2018 - Mar 2019 · 6 mos, Bengaluru, Karnataka, India

Prateek Verma - Deputy Manager (Senior Consultant) in Bengaluru | Tal