Headline

AI/ML Engineer - Shipping production agentic systems, distributed pipelines & LLM applications at enterprise scale · Open to global opportunities

About

I build AI systems that actually run in production — not demos. Currently at Accenture as an AI/ML Computational Science Analyst, where I architect and ship multi-agent LLM pipelines, distributed event-driven backends, and enterprise automation systems serving 400+ monthly users across Fortune 500 clients. My stack sits at the intersection of agentic AI and backend engineering: LangGraph, LangChain, Kafka, RabbitMQ, FastAPI, Django, PostgreSQL, Azure. I've implemented Raft consensus from scratch, built RAG pipelines, designed prompt engineering frameworks for multi-agent systems, and cut client delivery time by 3x through intelligent workflow automation. B.Tech in Computer Science & Artificial Intelligence from IIIT Delhi (CGPA: 8.17). I'm drawn to roles where I can own the full loop — frame the problem, build the system, ship it to real users, and improve it based on what production teaches me. Open to: AI/ML Engineering · Agentic Systems · Backend / Distributed Systems · Data Engineering · Forward Deployed Engineering · GenAI Platform roles LangGraph | Multi-Agent Systems | RAG | Kafka | FastAPI | PyTorch | Distributed Systems | Azure | GCP | Open to relocation

Location

Greater Delhi Area

AI stack

cursor

opencode

claudecode

Work experience

AI/ML Computational Science Analyst at Accenture, Jun 2025 - Present · 1 yr 3 mos, Gurugram, Haryana, India, • Architected a production LangGraph multi-agent system using ReAct reasoning and hierarchical delegation — orchestrating LLM tool-calling, layout analysis, and output validation end-to-end; reduced enterprise client delivery time by 3x over manual workflows. • Built a fault-tolerant, event-driven distributed pipeline on Azure (Kafka, RabbitMQ, Celery, PostgreSQL) adapting PSD, INDD, HTML, and PDF marketing assets across independent microservices — serving 400+ monthly enterprise users with zero data loss under peak concurrent load. • Integrated Adobe Firefly and LLM APIs (OpenAI, Claude) via Django/FastAPI microservices to power automated creative generation pipelines across multiple enterprise accounts. • Established automated PyTest regression suites, telemetry dashboards, and distributed fallback strategies across all agentic components — reducing production error rates by 40%. • Acted as technical point of contact for enterprise clients, translating business problems into agentic solutions and surfacing recurring product gaps back to engineering — directly influencing the internal roadmap.

Student Volunteer at National Service Scheme, Sep 2023 - Dec 2024 · 1 yr 4 mos, India

Intern at upSkill Campus, Jun 2023 - Jul 2023 · 2 mos, Remote, Developed and evaluated 4 ML models (Linear Regression, Decision Tree, Random Forest, Neural Network) to forecast Smart City traffic patterns from historical data, selecting the best-performing model through rigorous cross-validation. Performed end-to-end feature engineering and statistical analysis; produced visualisations of predicted traffic patterns by junction, day of week, and month to support data-driven urban planning decisions.

Campus Ambassador at Coding Ninjas, Aug 2022 - Nov 2022 · 4 mos, New Delhi, Delhi, India, Drove technical community engagement and developer outreach across IIIT Delhi; organised coding events and represented the platform to student engineers.

Education

Bachelor of Technology - BTech, Computer Science, Indraprastha Institute of Information Technology, Delhi

12th, Non Medical, CRPF Public School - India

Public links

github: https://github.com/Aryan-Dhull-Dev