Aakash Priyadarshi

Bibliography

Professional Summary

Software Engineer with hands-on experience building production-grade AI agents, full-stack applications, and LLM-powered systems. Proficient in designing scalable architectures that blend real-time performance, custom tooling, and human-like interaction. Skilled in building with Python, TypeScript (React), and deploying LLMs using frameworks like LangChain, RAG pipelines, and custom orchestration logic. Experienced in customer-facing development and rapid iteration across ambiguous problem spaces.

Education

• MSc in Computer Science – University of Liverpool (Distinction, 72%)
  • Awarded top grades in dissertation and coursework modules.
  • Scored 86.68 on COMP-517 Programming Fundamentals, ranked among top 8 students.
  • Dissertation: Building a Predictive Model for Stock Market Prices (Integrated advanced machine learning techniques).
• B.Tech in Electronics and Instrumentation Engineering – Galgotias College of Engineering and Technology (Merit).

Patents

1. Artificial Intelligence Integrated Robotic Military Suit (2024) - Shortlisted by DRDO for military use
2. User Interactive Assistive Learning Educational Content Generation AI Platform (2023)
3. Industrial Load Lifting Robotic Arm Exosuit (2021)

Achievements

• Ranked 4th in Bihar for the National English Olympiad during high school.
• Two consecutive national-level selections for Science Exhibitions.
• My projects, such as HIND: A WebRTC Video Conferencing Platform, received media coverage and public appreciation.
• DRDO Patent Recognition – Patent for AI Integrated Robotic Military Suit shortlisted by DRDO for military use.

Professional Experience

Full-Stack AI Engineer | Mercor | Remote, USA

February 2025 – Present

• Engineered internal developer tools for GPT evaluation workflows, ensuring secure and scalable production environments using Python (FastAPI) and React.js
• Built internal GPT evaluation pipelines using FastAPI + React, deployed via Docker on AWS Lambda and EC2
• Authored Python scripts to automate data collection and semantic scoring, supporting large-scale LLM testing
• Designed full-stack architecture supporting containerization, observability, and cross-team collaboration with AI researchers

Senior Software Developer & Technical Lead | Bihari Construction & Infra Pvt. Ltd. | New Delhi, India

July 2020 – September 2023

• Led development of internal web apps (React, Node.js) for team tracking, project operations, and finance
• Automated key business operations using REST APIs, WebSockets, and SQL data pipelines
• Delivered tools that improved response times and helped CS/Ops teams reduce routine tasks
• Owned sprint planning and code delivery with cross-functional teams

Extended Career Experience

September 2024 – February 2025

• Computer Science Professor | Wirral Met College, Birkenhead, UK
  Teaching Python programming and web development, fostering innovation among students

• Teaching Assistant | University of Liverpool, Liverpool, UK
  Guided undergraduate students in Software Engineering and Cybersecurity modules

• Remote Student Mentor Volunteer | Coursera, California, USA
  Mentored students globally in computer science and programming fundamentals

Publications and Research

• Open-Source Contribution: Core contribution to Astropy – Fixed FITS logical variable-length array storage bug (14+ years old). Merged PR with 500+ lines across 4 files, comprehensive testing, and maintainer collaboration. GitHub PR #18778
• Publication: Deep Learning Innovations in Automated Breast Cancer Detection with Integrated Ultrasound Datasets
• Authored research on AI-driven trading systems, published dissertations, and conference papers.
• Focus on AI hallucination mitigation, advanced robotics, and predictive analytics.

Skills

• Agent & LLM Tooling: LangChain, RAG pipelines, OpenAI APIs, Pinecone, Retrieval QA, prompt engineering
• Frontend: React.js, Next.js, TypeScript, Tailwind CSS, Shadcn
• Backend: Python (FastAPI, Flask), Node.js, REST APIs, WebSockets
• ML Deployment: Hugging Face Transformers, vector stores, structured prompt frameworks
• Cloud & DevOps: GCP, AWS, Docker, GitHub Actions, Cloud Run
• Database & Tools: Firebase, PostgreSQL, Redis, CI/CD, pytest, Playwright

Personal Projects

• Machine Learning Model to Forecast Company Stock Prices - Built a cloud-based web app using React.js, Python (Flask), and AWS with RESTful APIs for real-time data processing. Deployed on AWS with 3x performance improvement. GitHub
• Chatbot for Contextual Customer Interaction - Real-time chat system with Socket.IO, semantic search using Hugging Face & Pinecone, integrated OpenAI GPT-3.5, reducing data retrieval time by 40%.
• AI Car Parking Management System - Built with React.js, Node.js, WebSockets, and predictive models using Python, OpenCV, TensorFlow, reducing search times by 30%.
• Open-Source LLM Developer Chatbot SDK & Platform - Full developer SDK using Node.js, OpenAI GPT-4, and Pinecone with semantic search, embedded logging, and Docker deployment. GitHub

Vision

I aim to revolutionize AI systems through impactful research and innovative solutions, specializing in production-grade AI agents and LLM-powered applications. My goal is to continue pushing the boundaries of what’s possible with artificial intelligence while building scalable, customer-facing solutions that solve real-world problems.