Open to opportunities

Abhijeet Mandal

@abhijeetmandal

Message

I build evaluated LLM agents, NLP ranking systems, and cloud automation.

India

Message

What I'm looking for

I'm looking to build and evaluate practical AI and LLM systems, especially retrieval, ranking, and agent workflows, while applying my AWS, Docker, CI/CD, and DevOps experience to reliable production deployments.

I've built AI-agent evaluation tasks at Afterquery Experts, creating Docker-containerized Bash and Python environments that measure autonomous reasoning and end-to-end task completion.

In HireAI, I built an explainable candidate-screening engine using schema-constrained LLM extraction, bge-m3 retrieval, and LLM verification, improving NDCG@5 from 0.70 to 0.96. I also shipped PII redaction, EEOC adverse-impact audits, Google Workspace OAuth integrations, and 245 GPU-free GitHub Actions tests.

I've also deployed AWS cloud platforms and DevSecOps pipelines with Docker, EC2, EKS, Jenkins, GitHub Actions, Prometheus, and Grafana.

Experience

Work history, roles, and key accomplishments

Current

Software Engineer (Bash/Linux)

Current

Afterquery Experts

Nov 2025 - Present (11 months)

Designed and implemented tasks on the Harbor framework for AI agent evaluation, contributing a real-world benchmark scenario. Built Docker-containerized terminal tasks in Bash and Python, including structured instructions, environment setup, and verification scripts.

BASH Python Docker Linux Harbor AI Agent Evaluation