Headline

AI Engineer | Expertise in LLMs, Vector Databases, AI Agents, Cloud & DevOps | Building Intelligent Applications

About

IIT Kharagpur | Integrated M.Sc. Economics (5Y) | Data Science & AI/ML Enthusiast Turning problem statements into intelligent systems. Passionate about building end-to-end Data Science and Machine Learning solutions that solve real business problems, from predictive modelling and causal inference to production-ready AI applications. I love working at the intersection of business and modern AI, turning real-world problems into intelligent, scalable solutions. From exploring raw data and engineering smart features to training powerful models and deploying them in production. Open to opportunities in Data Science, Machine Learning Engineering, and AI Product roles. Let’s connect and build something meaningful!

Location

Bengaluru

AI stack

anthropic apis

claude

codex

cursor

hugging face

claudecode

Work experience

AI Engineer at OnFinance AI, Jun 2025 - Present · 1 yr 3 mos, Bengaluru, Karnataka, India, Building AI Agents for revolutionising risk, audit and compliance for BFSI. • Architected and led Agent Studio, a LangGraph-based agentic workflow engine featuring a connected file-system, dynamic tool registry, and skill-prompt routing, enabling configurable, scalable AI pipeline orchestration across multiple product verticals; comparable in design to Codex or ChatGPT Agent Builders. • Engineered high-volume LLM, RAG, and document intelligence pipelines processing 100M+ tokens/month using LiteLLM, AWS Bedrock (Claude 4.5/4.6 Sonnet, Haiku 4.5), OCR, PyPDF, FusionQuery search, and re-ranking for large-scale financial and regulatory document ingestion. • Owned AWS Kubernetes infrastructure for 20+ enterprise clients, managing shared clusters with dedicated node pools for databases, application services, and LLM inference via Bifrost; operated production data systems across MongoDB, Redis, Qdrant, RabbitMQ, and Trino with CI/CD via GitHub Actions and Argo CD. • Drove end-to-end client delivery across 20+ enterprise accounts, including NSE, BSE, Axis Finance, BharatPe, PayU Finance, and FedFina etc., covering solution finalisation, tenant provisioning, integration planning, infra sizing, network provisioning, and sign-off across SaaS and on-premise deployments on AWS. • Secured $260K+ in infrastructure credits — $10K in AWS credits for POC deployments and $250K via Cloudflare Startup Program, engaging directly with Cloudflare engineering stakeholders on organizational-level integration planning and implementation. • Contributed to securing $4.2M Pre-Series A from PeakXV Partners, deeply involved in early product development including agentic workflows over Indian regulatory data and clause-level and metadata-level RAG pipelines over regulatory communications.

LLM Trainer at Outlier, May 2025 - Jun 2025 · 2 mos, Remote

LLM Trainer at Soul AI, Oct 2024 - Dec 2024 · 3 mos, Remote, • Assessed outputs from multiple LLMs in both STEM and non-STEM fields, focusing on accuracy and correctness • Reviewed code execution of AI systems, evaluating tool calls, function parameters, and API integrations • Designed and optimized prompts using techniques such as zero-shot, few-shot, and chain-of-thought prompting

AI Intern at Schneider Electric, May 2024 - Jul 2024 · 3 mos, Bengaluru, Karnataka, India, - Developing a knowledge bot to support developers. - Exploring GenAI solutions for knowledge management, testing, coding, integrations, and other AI-related areas. - Emphasize on AI-driven knowledge management and its applications within the organization

Education

Master of Science - MS, Economics, Indian Institute of Technology, Kharagpur

Delhi Public School Patna

Public links

github: https://github.com/HimanshuKumar17052001