Sanjay Charitesh Makam — Software Engineer & ML Researcher

Education

Account Origin

Aug 2022 — Dec 2026

Rochester Institute of Technology

B.S. Computing and Information Technologies

Concentration: Database & Web Development · Cybersecurity, Networking, OS coursework

★ Dean's List — Spring 2025, Fall 2025 & Fall 2026 (3 consecutive semesters)

3.70

GPA / 4.0

Experience

Transaction History

Period Description Type

May 2026
— Present

AI & Data Researcher City of Rochester · RIT · Hybrid

Collaborative project with Rochester Police Dept. & U.S. Dept. of Justice on body-worn camera footage AI analysis Applies ML, Computer Vision, and NLP to improve automated insights and law enforcement decision-making Developing interdisciplinary AI-driven framework for transparency and training support

ACTIVE

Research
Internship

Jan 2026
— May 2026

Research Assistant — VLA Security & Adversarial ML Rochester Institute of Technology · On-site

Investigated adversarial attack surfaces in 3+ categories within Vision-Language-Action models using PyTorch Trained on behavioral patterns across robotic perception and multimodal decision pipelines Tuned classification thresholds and learning rates across 10+ configurations for visual injection defense

RESEARCH

Academic
Research

Jan 2026
— May 2026

Teaching Assistant — NSSA 241 (Routing & Switching) Golisano College, RIT · On-site

TA for Intro to Routing and Switching — networking protocols, Wireshark, and lab configurations Supported students through hands-on network troubleshooting and subnet exercises

TEACHING

Networking TA

Aug 2025
— Dec 2025

Teaching Assistant — ISTE 140 (Web Development) Golisano College, RIT · On-site

Mentored 60+ students in full-stack development, frontend, and systems fundamentals Graded 200+ assignments; improved lab completion rates through one-on-one debugging

TEACHING

60+ Students

Apr 2025
— Aug 2025

Research Intern — Cybersecurity & ML RIT Cybersecurity Dept. (Prof. Dr. Naresh Kshetri)

Built malware detection on EMBER-2018 (1.1M samples) using RNN-LSTM + Transformers on PyTorch with GPU acceleration Engineered histogram + byte entropy pipelines; 97% LSTM accuracy, 99% Transformer accuracy Published at ACM SIGCITE 2025 · IEEE JCM 2026 poster presentation — won $100 prize & certificate

RESEARCH

ACM Published
IEEE Prize

May 2025
— Aug 2025

Software Engineer Intern Criyaa Inc. · California · Remote

Built full-stack e-signature platform: FastAPI backend + React frontend + AWS S3 + Auth0 + Dropbox Sign API Designed presigned URL workflows for secure file upload — maintained 100% uptime across all deployments Reduced document turnaround time by 25% through Agile sprint collaboration (Jira, Taiga, GitHub Actions CI/CD)

INDUSTRY

−25%
Turnaround

Projects

Key Investments

SpendSwitch

LIVEYC S26 Applicant

Dec 2025 — Present

Fintech SaaS that analyzes bank statements and tells users which card gives best cashback with exact dollar estimates per category 40+ users across 7 countries, 1,200+ transactions tracked — built solo end-to-end in 3 months Dynamic model-shifting: lightweight model + rules first, escalates to GPT-4 only on failure — reduces API cost Category caching, cross-card sub-category analytics, spending health score, Stripe + Razorpay billing on Google Cloud Run

40+
Users

7
Countries

1,200+
Transactions

30+
APIs

REACT.JSNODE.JSEXPRESS MONGODBOPENAI GPT-4STRIPE RAZORPAYGCPJWT CHART.JSFIREBASE

spendswitch.com →

Beehive Health Monitoring System

RIT

Aug 2025 — May 2026

Remote IoT monitoring using LoRa for real-time beehive temperature tracking and hive health visualization Full-stack dashboard: multi-hive management, viewer access control, graphical real-time + historical sensor readings OTP auth, email alerts for anomalies, MySQL on Ubuntu VM with Nginx, hardware-to-cloud-to-web pipeline

NODE.JSMYSQLJWTLORA/IOTNGINXUBUNTU SERVERREACT.JS

AI Virtual Assistant

AI/ML

Aug 2023 — Sep 2023

Conversational AI assistant using OpenAI APIs + LangChain for question-answering and multi-turn dialogue Integrated Hugging Face NLP models and PlayHT voice cloning for real-time speech generation Deployed via Gradio on Google Colab with API-driven real-time responses

OPENAI APILANGCHAINGRADIOHUGGING FACEPLAYHTPYTHON

Live Demo →

iSchool Information Portal

RIT

Jan 2025 — May 2025

Full-stack web app: React.js frontend + ASP.NET MVC (C#) backend consuming RESTful APIs for degrees, minors, faculty, courses Courses-on-demand feature, animations, responsive UI, reusable component architecture

REACT.JSASP.NET MVCC#REST APIsJAVASCRIPT

View Project →

TPC Data Warehouse & ETL Pipeline

RIT

Oct 2025 — Nov 2025

Dimensional data warehouse with star schema for The Product Company — ETL across customers, products, invoices Data cleansing, anomaly detection, source-to-target mappings, indexing optimization for BI reporting

MYSQLSASPENTAHO KETTLEETLSTAR SCHEMA

FundGoodDeeds — Financial Tracking App

Aug 2025 — Dec 2025

Java app with Observer, Factory, Composite, Facade, Mediator design patterns — 30% data-processing efficiency gain CSV-based persistent storage, daily ledger, financial analytics, multi-view GUI with full MVC separation

JAVAOOPSWING/JAVAFXDESIGN PATTERNSMVC

Dynamic Multi-Level Form Builder

RIT

Jan 2025 — May 2025

JavaScript DOM manipulation for dependent multi-level dropdown menus, form validation, animations Cookies/localStorage integration, browser compatibility, requestAnimationFrame() animations, reusable architecture

JAVASCRIPTHTML5CSS3DOMLOCALSTORAGE

View Project →

Research

5 Papers · 1 Published · 4 In Progress

secRNNlong: Malicious Code Classification using RNNs with LSTM and Transformers for improved Cybersecurity

Makam, S.C. et al. · EMBER-2018 (1.1M samples) · 97% LSTM · 99% Transformer · Sacramento, CA · IEEE JCM 2026 poster · $100 prize

ACM SIGCITE 2025

Post-Quantum Cryptography for Securing AI Model Weights at Scale

ML-KEM-768 + ML-DSA · PQC adds 0.003% latency overhead · Chunked secure inference · SHA3 vs BLAKE3 benchmarks

Target: USENIX 2026

Detecting Prompt Injection Attacks in Multi-Agent AI Systems

LLM-to-LLM injection detection classifier · NLP + RNN/Transformer pipeline · Co-authored

Target: USENIX 2026

Deep Learning for Fetal Brain MRI Segmentation and Structural Analysis

Automated fetal brain tissue segmentation using transformer-based deep learning · Prenatal neuroimaging · PyTorch · Conference TBD

In Progress

Adversarial Knowledge Injection in RAG Systems: How Much Poison Controls the AI?

Quantifying document-level poisoning thresholds in Retrieval-Augmented Generation · Attack surface analysis · NLP + classifier pipeline

Target: USENIX 2026

Skills

Asset Portfolio

Languages

PythonJavaScriptTypeScript JavaCC# (.NET) SQLPHPHTML5 CSS3Bash / Shell

Frameworks & Libraries

React.jsNode.jsExpress.js FlaskFastAPIASP.NET MVC PyTorchPandasLangChain GradioHugging Face

AI / ML

OpenAI GPT-4Deep Learning NLPRNN / LSTMTransformers Adversarial MLComputer Vision AI AgentsMulti-Agent Systems RAGPrompt Engineering LLM Orchestration

Cloud & DevOps

AWS S3Google Cloud Run DockerGitHub Actions FirebaseNginx CI/CDUbuntu Server

Databases & Data Engineering

MongoDBMySQL SQLiteMSSQLOracle ETL PipelinesStar Schema SAS StudioPentaho KettleWeka

Backend, Auth & APIs

REST APIsJWT OAuth 2.0Auth0 StripeRazorpay Dropbox SignWebhooks MVC ArchitectureSOAP / XML

Networking & Systems

Active DirectoryDNS / DHCP pfSense / FirewallVMware Windows Server 2025Rocky Linux Apache httpdSSL / TLS Samba / NFS / FTPRAID SMTP / IMAPWireshark Virtual LANCron Jobs Linux Sysadmin

Tools, Design & IoT

FigmaJira TaigaGit / GitHub Agile / ScrumKanban LoRa / IoTRaspberry Pi Google ColabJupyter

Systems & Networking Labs

Enterprise Infrastructure Experience

LAB 01

Virtualization & Network Setup

VMwarepfSenseNAT/LANWin Server 2025Rocky LinuxStatic IP

LAB 02

Active Directory, DNS & DHCP

AD DSDNS ServerDHCPGroup PolicyDomain ControllerUser Mgmt

LAB 03

Version Control & Git

GitLocal/Remote ReposCommit/Push/PullBranching

LAB 04

Partitioning & RAID

MBR/GPTRAID 1 & 5mdadmfstabRAID RecoveryFile Systems

LAB 05

File Services

FTP/vsftpdSambaNFSRsyncSELinuxCross-platform Sharing

LAB 06

Web, Email & Automation

Apache httpdSSL/TLSVirtual HostsSMTP/IMAPCron JobsRemote Logging

Upcoming

Scheduled Deposits

STEMforAll 2026 — University of Rochester

July 27 – August 7, 2026 · Hylan Building, UR · Neural Network Pruning Research (Group A2)

Two-week in-person research workshop with University of Rochester faculty. Investigating useless neuron detection and neural network pruning using ResNet-18 — identifying which neurons can be removed without sacrificing model accuracy. Extends existing deep learning and adversarial ML research background.

Content Creator

Teaching the Internet