Aug 2024 - May 2026
Master of Science, Computer Science
University of Colorado Boulder, USA
GPA: 4.0 / 4.0
// 01 — About
About
Machine learning engineer pursuing an M.S. in Computer Science at CU Boulder, focused on building GenAI and backend systems that are practical, reliable, and production-ready.
AI/ML Engineer
Building reliable GenAI, retrieval, and backend systems.
I build agentic workflows, RAG pipelines, and ML services with LangGraph, FastAPI, and modern cloud tooling.
Phone
+1 7202091106
Location
Boulder, CO, United States
Degree
MS in Computer Science, University of Colorado Boulder
Email
tejusp.us@gmail.com
Currently working as an AI Engineer at Mulberry LLC, where I built a vision-guided desktop automation agent in Electron, designed a two-stage grid-aiming pipeline with per-display auto-calibration for pixel-accurate click targets, and architected an in-process MCP-style tool layer and agentic loop using native function-calling.
Before that, I was a Machine Learning Engineer at Treosoft IT Solutions and a Software Engineer Intern at Ford Motor Company, where I shipped recommendation systems for 100K+ daily transactions, built Python and SQL ETL pipelines, developed LangChain-based RAG systems with FAISS and Cohere Rerank, and created internal search tools powered by OpenAI embeddings.
I also completed research at National Institute of Technology Puducherry, with work published through IEEE, Springer, and Computational Biology and Chemistry. Highlights include a Best Paper Award at IEEE IConSCEPT 2023, AWS GameDay Winner 2025, and 2nd Place at the CU Boulder Expo 2026.
Got an interesting opportunity? Get in touch.
// 02 — Stack
Tech Stack
Technologies I work with daily
AI & Machine Learning
PyTorch
TensorFlow
scikit-learn
OpenAI
LangChain
LangGraph
HuggingFace
OpenCV
Cloud & DevOps
AWS
GCP
Docker
Kubernetes
Cloud Run
Vertex AI
Data & Databases
PostgreSQL
MongoDB
FAISS
ChromaDB
Pandas
NumPy
Languages & Frameworks
Python
C++
JavaScript
TypeScript
FastAPI
Git
// 03 — Resume
Resume
Last Updated on April 2026
April 2026 Snapshot
AI/ML engineer building production-ready GenAI, retrieval, and backend systems.
Pursuing a Master of Science in Computer Science at the University of Colorado Boulder while building LLM applications, RAG systems, FastAPI backends, and applied machine learning pipelines with an emphasis on real deployment tradeoffs.
MS in CS
University of Colorado Boulder · 4.0 / 4.0 GPA
Industry Experience
ML engineering, RAG, retrieval, recommender systems, and internal AI tools
Core Focus
LangGraph, FastAPI, LLM apps, vector search, cloud deployment, and applied ML
Education
Academic Foundation
Dec 2020 - May 2024
Bachelor of Technology, Computer Science Engineering
National Institute of Technology Puducherry, India
GPA: 8.5 / 10
Experience
Professional Work
June 2026 - Present
Remote, US
AI Engineer
Mulberry LLC
- Built a vision-guided desktop automation agent in Electron to complete natural-language goals on any native app.
- Designed a two-stage grid-aiming pipeline with per-display auto-calibration to resolve pixel-accurate click targets.
- Architected an in-process MCP-style tool layer and agentic loop using native function-calling.
May 2025 - Aug 2025
Remote, US
Machine Learning Intern
The Sprouting Company
- Built a labeled dataset of sprouted seed images for mold vs root-hair detection.
- Compared a custom CNN, fine-tuned EfficientNet-B2, and prompt-engineered GPT-5 for fungal detection, with EfficientNet-B2 reaching 93.6% mold accuracy.
- Compared model accuracy, API cost, and AWS deployment tradeoffs to recommend the best option for production use.
Aug 2023 - Jul 2024
Bangalore, India
Machine Learning Engineer
Treosoft IT Solutions
- Deployed a recommendation engine for 100K+ daily transactions using XGBoost, increasing net sales by 15%.
- Built Python and SQL ETL pipelines and trained a churn model with 20+ features that reached an AUC-ROC of 0.85.
- Built a RAG system with LangChain that improved internal benchmark accuracy by 35%.
- Combined FAISS search with Cohere Rerank and caching to reduce average query latency by 50%.
May 2023 - Jul 2023
Chennai, India
Software Engineer Intern
Ford Motor Company
- Built an internal knowledge search tool using OpenAI embeddings and FAISS to help engineers query legacy PDFs.
- Built Python and Pandas pipelines for ALM data that cut manual processing time from 2 hours to 1 minute.
- Created Tableau dashboards for 20+ teams to track model performance, data drift, and engineering bottlenecks.
Apr 2022 - Apr 2023
Puducherry, India
Research Intern
National Institute of Technology Puducherry
- Conducted machine learning research in document classification, medical imaging, and protein localization.
- Co-authored papers published through IEEE, Springer, and Computational Biology and Chemistry.
// 04 — Projects
Projects
Selected projects across agentic systems, full-stack AI products, and applied machine learning.
Featured Work
Agentic systems, full-stack AI products, and applied ML research.
Flagship Build
AI Personal Trainer
Injury-aware workout planning app built with LangGraph, FastAPI, and Streamlit. Trainer and physiotherapist agents review plans before the final program reaches the user.
Product Build
Pantry to Plate
Food-waste reduction app that turns fridge images into ingredient-aware recipes using FastAPI, PostgreSQL, Cloud Run, and Vertex AI.
Systems Work
Autonomous Refactoring Agent
Local code refactoring agent with a self-healing loop that plans changes, runs tests, validates AST safety, and retries failed edits automatically.
R&D
Multi-Agent RAG Research
Experimental repository for exploring multi-agent retrieval flows, document reasoning patterns, and research-focused orchestration ideas around RAG systems.
Applied ML
MoodyTunes
Music mood modeling pipeline using LastFM tags, audio-derived features, and deep learning models to recommend songs based on emotional tone.
Full-Stack AI
SeeMyRace
Google OAuth and Drive-based race photo platform with sequential ML processing, selfie and bib-based athlete search, EXIF GPS mapping, and admin moderation for flagged images.
// 05 — Papers
Papers
Peer-reviewed publications and conference proceedings.
Gene Ontology graph embeddings with Dynamic Thresholding based Deep Neural Networks for Multi-label protein subcellular localization prediction
Investigating Feature Extraction and Classification Algorithms for Effective Lung Disease Detection Using Chest X-Ray Images
Resume Classification Using ML Techniques
Text Feedback Classification using Machine Learning Techniques
Potato Disease Classification Using Diverse Feature Extraction Methods and Machine Learning Models
// 06 — Contact
Contact
If you are hiring for AI/ML, GenAI, RAG, or backend engineering work, I would be glad to connect.