Looks like you have JavaScript disabled. For the full Twine experience, you will need to re-enable it.

AI Engineer with 4+ years of experience designing and deploying production-grade GenAI, RAG, and Agentic AI systems. Proficient in GPT-4, LangChain, LangGraph, FastAPI, and AWS infrastructure, with a focus on building reliable, scalable AI solutions. Experienced in multi-agent workflows, LLM/RAG evaluation frameworks, and production deployment (MLOps, CI/CD, Docker, Kubernetes). Adept at reducing hallucinations, enforcing guardrails, and delivering enterprise-grade AI systems with high uptime and low latency.…AI Engineer with 4+ years of experience designing and deploying production-grade GenAI, RAG, and Agentic AI systems. Proficient in GPT-4, LangChain, LangGraph, FastAPI, and AWS infrastructure, with a focus on building reliable, scalable AI solutions. Experienced in multi-agent workflows, LLM/RAG evaluation frameworks, and production deployment (MLOps, CI/CD, Docker, Kubernetes). Adept at reducing hallucinations, enforcing guardrails, and delivering enterprise-grade AI systems with high uptime and low latency.

Ashok Kumar

AI Engineer, Data Scientist, Programmer, +2





AI Engineer with 4+ years of experience designing and deploying production-grade GenAI, RAG, and Agentic AI systems. Proficient in GPT-4, LangChain, LangGraph, FastAPI, and AWS infrastructure, with a focus on building reliable, scalable AI solutions. Experienced in multi-agent workflows, LLM/RAG evaluation frameworks, and production deployment (MLOps, CI/CD, Docker, Kubernetes). Adept at reducing hallucinations, enforcing guardrails, and delivering enterprise-grade AI systems with high uptime and low latency.…AI Engineer with 4+ years of experience designing and deploying production-grade GenAI, RAG, and Agentic AI systems. Proficient in GPT-4, LangChain, LangGraph, FastAPI, and AWS infrastructure, with a focus on building reliable, scalable AI solutions. Experienced in multi-agent workflows, LLM/RAG evaluation frameworks, and production deployment (MLOps, CI/CD, Docker, Kubernetes). Adept at reducing hallucinations, enforcing guardrails, and delivering enterprise-grade AI systems with high uptime and low latency.

Available to hire

AI Engineer with 4+ years of experience designing and deploying production-grade GenAI, RAG, and Agentic AI systems. Proficient in GPT-4, LangChain, LangGraph, FastAPI, and AWS infrastructure, with a focus on building reliable, scalable AI solutions.

Experienced in multi-agent workflows, LLM/RAG evaluation frameworks, and production deployment (MLOps, CI/CD, Docker, Kubernetes). Adept at reducing hallucinations, enforcing guardrails, and delivering enterprise-grade AI systems with high uptime and low latency.

Skills

Experience Level

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Intermediate

Intermediate

Work Experience

AI Engineer at Cognizant

May 1, 2025 - Present

Architected production RAG pipelines using GPT-4 and Azure OpenAI via LangChain with Pinecone to enable enterprise document Q&A with sub-2s response latency. Designed and deployed agentic AI workflows with LangGraph and tool calling, reducing manual analyst intervention by 45%. Built real-time inference APIs using FastAPI on AWS ECS and API Gateway supporting 500+ concurrent users with 99.9% uptime. Implemented LLM/RAG evaluation frameworks using RAGAS and TruLens, reducing hallucination rates by 40% via automated quality benchmarks. Enforced guardrails to detect toxic outputs, prompt injections, and off-topic responses, improving AI safety compliance across deployed models.

AI & ML Associate Engineer at Infosys

August 1, 2020 - October 1, 2023

Developed and evaluated open-source LLM solutions (Flan-T5) using LangChain for document summarization and information extraction on unstructured text datasets. Integrated Llama 2 and Mistral models with vector embedding pipelines (Hugging Face Transformers) to build context-aware Q&A, reducing manual review time by 35%. Designed and trained supervised classification models (Logistic Regression, Random Forest, XGBoost) using scikit-learn with cross-validation to optimize F1. Built ETL pipelines with PySpark to process high-volume structured and unstructured datasets, reducing data preparation time by 25% for training workflows.