Headline

AI Systems Architect | OSS Contributor | GenAI | RAG | Agentic Systems | FastAPI | AWS | Azure | Financial AI | Distributed Systems

About

I am an AI Engineer specializing in Agentic Workflows, Model Context Protocol (MCP), and Distributed Systems. My focus is on moving LLMs out of the chat window and into reliable, governed enterprise execution. Currently at PL Capital, I architect production AI platforms that transform unstructured financial data and audio into structured intelligence, recently reducing manual verification efforts by 85-90%. I don't just build wrappers; I engineer autonomous systems that are safe, observable, and fast. A few highlights of what I build: Autonomous Agents: Engineered FinSight, a 5-tool MCP-native financial research agent orchestrating parallel fetch and analysis, cutting research workflow time by 97%. AI Governance & Security: Published a framework-agnostic governance SDK to PyPI and built a zero-trust gatekeeper for Kubernetes infrastructure operations. High-Performance Infrastructure: Developed a distributed LoRA training orchestrator with heartbeat-based failover and a Go-based distributed KV cache recovering from master termination in 300ms. Core Stack: Python, Go, LangGraph, FastAPI, PyTorch, ChromaDB, and OpenTelemetry. Let's connect if your team is building the next generation of reliable AI agents.

Location

Gurugram

AI stack

antigravity

claude code

codex

hugging face

OpenAI Agents SDK

OpenRouter

Work experience

Artificial Intelligence Engineer at PL Capital Group (Prabhudas Lilladher), Aug 2025 - Present · 1 yr 2 mos, Gurugram, Haryana, India

Artificial Intelligence Engineer at PL Capital (PLindia), Aug 2025 - Present · 1 yr 2 mos, Built an end-to-end compliance verification system (Python, asyncio, Oracle DB) processing trade records and supporting documents across network shares; deployed a custom ONNX image-classification model (98.82% accuracy) for local inference, automated multi-format ingestion pipelines, and implemented validation layers including PII masking, chronology analysis, completeness scoring, and entity matching to generate analyst-ready compliance reports.

Artificial Intelligence Engineer at PL Capital (PLindia), Aug 2025 - Present · 1 yr 2 mos, Delivered a TypeScript full-stack platform (React/Vite, Express, 46 REST endpoints, 14+ MySQL tables via Drizzle ORM) with role-based access control (admin/analyst/associate), a 4-step extraction wizard, and Prometheus API-latency metrics.

Artificial Intelligence Engineer at PL Capital (PLindia), Aug 2025 - Present · 1 yr 2 mos, Architected a production document-intelligence platform (Python, FastAPI, asyncio) extracting structured financial metrics from unstructured reports through concurrent OCR and vision-model pipelines; integrated FAISS-powered semantic retrieval with confidence scoring, SHA-256 extraction caching, and request-level cost/latency observability; implemented financial validation rules, unit normalisation, and temporal eligibility checks to ensure enterprise-grade extraction accuracy.

AI Intern at SKV (Studiokon Ventures Private Limited), Jun 2025 - Aug 2025 · 3 mos, Gurugram, Haryana, India

Software Engineer Intern at StudioKon Ventures, Jun 2025 - Aug 2025 · 3 mos, Built a full-stack procurement platform (TypeScript, React, Express, NeonDB/PostgreSQL, Drizzle ORM) with composite index optimisation cutting p95 query latency by ∼40%; integrated AWS S3 and WhatsApp webhook notifications; deployed via GitHub Actions CI.

Tech Head at Ragam National Institute of Technology Calicut, Sep 2024 - Feb 2025 · 6 mos, NIT Calicut

Education

Bachelor of Technology - BTech, Electronics and Communications Engineering, National institute of technology Calicut

Senior Seccondary, St Joseph's Convent School Varanasi

Public links

github: https://github.com/Deadeye102000

portfolio-website-deadeye4.vercel.app: https://portfolio-website-deadeye4.vercel.app/