Deepak Thirukkumaran
AI EngineerAI is analyzing your overall score…
Identifying your key strengths…
Evaluating your skill match against the job requirements…
Assessing your cultural and operational fit
AI is analyzing your overall score…
Identifying your key strengths…
Evaluating your skill match against the job requirements…
Assessing your cultural and operational fit
AI Engineer with less than a year in Computer Vision, NLP, and LLMs
Deepak Thirukkumaran is an aspiring AI/ML Software Engineer with 1 year of experience, specializing in computer vision, natural language processing, and large language models. With a background in developing scalable AI solutions using FastAPI, PostgreSQL, and modern ML frameworks, Deepak has contributed to projects spanning manufacturing optimization, multimodal 3D object detection, and intelligent bid platforms, demonstrating strong skills in full-stack development and cloud infrastructure.
Shiv Nadar University, Chennai
B.Tech · Artificial Intelligence & Data Science
August 1, 2021 – June 30, 2025
ORCA Digital Lean Solutions
AI/ML Software Engineer
November 1, 2025 – Present
Detroit, Michigan, United States
CreatorOS by DRPCRD
AI Engineer Intern
January 1, 2025 – July 1, 2025
London, England, United Kingdom
IIT Hyderabad
Research Intern
May 1, 2024 – September 1, 2024
Hyderābād, Telangana, India
Sports Mechanics
Data Analyst Intern
December 1, 2023 – February 1, 2024
Chennai, Tamil Nadu, India
Tata Consultancy Services (TCS)
Computer Vision Intern
May 1, 2023 – July 1, 2023
Chennai, Tamil Nadu, India
Tendering AI - Bid Intelligence Platform
June 1, 2026 – Present
Designed a multi-agent AI system for Oil & Gas tender automation, orchestrating 8 agents across qualification, BOM generation, supplier evaluation, and proposal assembly. Implemented HITL approval workflows and a Gemini-powered qualification agent for automated Go/No-Go decisions.
View ProjectMedFusion: Vision-Language Framework for Radiology
June 1, 2026 – Present
Multimodal radiology VQA and report generation system combining ConvNeXt, fine-tuned CLIP (ViT-32B), and Med-LLaMA with residual feature fusion. 76.3% accuracy on SLAKE, 72.8% on PMC-VQA benchmarks.
View ProjectFintech Intelligence Pipeline
June 1, 2026 – Present
AI-powered stock analysis platform with live market data (Polygon.io), LLaMA 3.2 sentiment/risk analysis, and automated investment insights. Deployed with Flask APIs, React dashboard, PostgreSQL, and Railway CI/CD.
View ProjectMiniGPT: Transformer Language Model from Scratch
June 1, 2026 – Present
22M parameter GPT implementation: token embeddings, multi-head self-attention, positional encoding, feed-forward layers. Training, generation, and top-k/nucleus sampling. 40 perplexity on WikiText-2.
IBM AI Developer
IBM
June 1, 2026 – Present
IBM Rag and Agentic AI
IBM
June 1, 2026 – Present
AWS AI Practitioner
AWS
June 1, 2026 – Present
Cultural Fit Analysis
The candidate's project diversity, spanning Fintech, Healthcare, Oil & Gas, and Manufacturing, indicates adaptability and a broad interest in applying AI across different industries. Their involvement in both academic research (IIT Hyderabad) and industry internships (TCS, CreatorOS, ORCA) shows a balanced approach to learning and application. The range of technologies used (PyTorch, HuggingFace, LangChain, Docker, AWS, GCP) suggests an openness to new tools and a continuous learning mindset, which aligns well with an innovative AI engineering culture. The personal projects demonstrate initiative and passion beyond academic requirements.
Soft Skills & Operational Fit
The candidate demonstrates strong problem-solving skills through their project descriptions, tackling complex challenges like multi-agent orchestration and multimodal data fusion. Their experience in various internships and personal projects suggests a proactive and self-driven work attitude. The descriptions imply an ability to work in cross-functional teams, especially in the context of deploying full-stack AI solutions. However, direct evidence of stress handling or explicit team collaboration skills is not provided.