RAG
Featured
01
Available for AI/ML Engineering & Data Science roles
Divyaansh Vats
AI/ML Engineering & Data Science · Building production-ready intelligent systems
I design and ship production-ready AI systems, grounded LLMs, retrieval pipelines, autonomous agents, and computer vision, that deliver measurable impact in real enterprise environments.
PyTorch
RAG
Agents
Computer Vision
LLMs
Education
Education
Foundations in mathematics, computing, and analytical thinking.
01
2022 - 2026CGPA 8.22 / 10
B.Tech in Mathematics and Computing
Rajiv Gandhi Institute of Petroleum Technology
Jais, Amethi, Uttar Pradesh
Minor in Business Analytics
02
2020 - 202291.2%
Senior Secondary (AISSCE, CBSE)
Kensri School and College
Bengaluru, Karnataka
Class XII, Science
HQ
HyperQuark Intelligence Lab
Research, Dynamic Career Intelligence Graph · 2026
- Contributing to the Dynamic Career Intelligence Graph research track
- Aligned with the lab's evolving research direction on career-signal modeling
- Emphasis on structured problem-solving and system-level reasoning
- Focused on developing tangible research outputs, not just prototypes
ResearchGraph ReasoningLLMsSystem Design
A8
Algo8.AI
Data Engineer Intern · Remote · Feb 2026 to Jun 2026
- Architected production ETL pipelines migrating data from EC2 to S3 for NeoStats analytical models
- Built large-scale feature engineering and batch workflows in PySpark for downstream ML readiness
- Enabled event-driven MLOps using AWS Lambda + EC2, reducing pipeline debugging turnaround
PythonPySparkAWS S3LambdaEC2ETL
FX
Fidrox Technologies
Data Science Intern · Bengaluru · May to Aug 2025
- Reduced anomaly verification time by 40% by automating analysis of 31,505 access logs
- 95% precision on security anomaly classification across 1,000+ critical incidents
- Built AccessAI, end-to-end pipeline flagging abnormal user behavior in enterprise access logs
PythonPyTorchAutoencodersAnomaly Detection
SP
Spatialty.AI
Machine Learning Intern · Bengaluru · May to Jul 2024
- Improved structural feature extraction on satellite imagery with Gabor filters + Multi-Otsu segmentation
- Accelerated vehicle annotation throughput via Make-sense.ai object detection workflows
OpenCVGabor FiltersSegmentationCV
RG
RGIPT, Prof. Rohit Bansal
Undergraduate Research Intern · Feb to Mar 2024
- Designed end-to-end ML workflows covering preprocessing, features, training, validation
- Evaluated model performance across configurations and documented findings for academic review
Pythonscikit-learnStatistical Modeling
Selected Work
Projects
A few systems I've built end-to-end, featured work first.
CV
Featured
02
Audit
Featured
03
AuditFlow AI
AI-powered platform that streamlines enterprise financial auditing, automates transaction validation, detects revenue leakage, flags anomalous financial patterns, and assists auditors with real-time insights. Built during the Lyzr Agentathon using LLMs and agentic workflows.
PythonLLMsAI AgentsRAGFinancial Analytics
CNN
04
CNN Architecture Benchmarking on CIFAR-10
Benchmarked three CNN architectures on CIFAR-10: baseline CNN reached 77.13% test accuracy, hyperparameter tuning added a 21.10pp lift, and an EfficientNetB0 transfer-learning variant hit 99.31% validation accuracy at low compute cost.
PyTorchEfficientNetB0MobileNetV2CIFAR-10CNN
FinML
05
ML-Based Return Predictors & the Spanning Controversy
Empirical study applying machine learning to micro-finance return prediction, testing whether ML factors span traditional linear predictors. Covers feature engineering on financial time series, cross-sectional model evaluation, and interpretation of spanning tests.
Pythonscikit-learnFinancial MLTime SeriesResearch
Toolkit
Skills & Stack
Grouped by discipline, the tools I reach for when shipping.
Programming
PythonC++JavaScriptTypeScriptSQLBash
AI & GenAI
LLMsRAGAI AgentsLangChainLangGraphLlamaIndexOpenAIAnthropicPrompt Engineering
Computer Vision
OpenCVYOLOPyTorch VisionMediaPipeImage Segmentation
Cloud
AWSS3LambdaGlueETL PipelinesDocker
Frameworks
PyTorchTensorFlowscikit-learnFastAPIStreamlitFlask
Databases
PostgreSQLMongoDBPineconeFAISSChromaRedis
MLOps
MLflowWeights & BiasesGitHub ActionsCI/CDModel Monitoring
Recognition
Achievements
Wins from national contests, hackathons, and competitive programming.
Top 1% Nationally
ALLEN SOPAN 2023, among 100,000+ participants
2nd Runner-Up
AlgoUniversity Graph Theory Programming Camp
Top 10 of 1,000+ Teams
Kode Current Hackathon
Ranked 38 / 500+ Builders
Lyzr Agentathon
3rd Place
Hacktoberfest Hackathon 2025
700+ CP Problems Solved
400+ CodeChef · 300+ LeetCode
About Me
I enjoy building production-ready AI systems that solve real-world problems using LLMs, RAG, AI Agents, Machine Learning, and Computer Vision. I enjoy learning new technologies, conducting research, and solving challenging engineering problems.