Skip to main content
Imani GadAtlanta, GA

Projects

Home

Projects

Product engineering, developer tooling, security investigation, and machine learning research.

Primary Projects

Neighborly marketplace home screen from its browser smoke-test artifactMarketplace / product

Neighborly

In Development

Full-Stack / Backend / Product Engineering

A request-first local marketplace for neighbors to discover and exchange housing, services, jobs, and community listings.

A React and TypeScript client integrates with a NestJS modular API; Prisma models the PostgreSQL-backed marketplace. Authentication and trust/safety workflows are implemented in dedicated backend modules.

ReactTypeScriptNestJSPostgreSQLVitePrisma
DevDash engineering health dashboardDeveloper activity / analytics

DevDash

Live

Developer Tooling / Analytics / Integrations

A developer productivity tool that turns GitHub repository, pull-request, issue, review, CI, and commit activity into action items, timelines, focus analysis, and weekly summaries.

GitHub OAuth and server-side GitHub API sync feed normalized repository, pull-request, issue, review, CI, and commit signals into durable Prisma-backed activity and sync snapshots. The application derives ranked action items, timelines, focus analysis, and grounded weekly summaries from that data.

Next.jsTypeScriptNextAuthPrismaGitHub APIOpenAI API
Investigation / evidence

AI Security Investigator

Prototype

Security Engineering / Applied AI

A CCTV investigation system for uploading footage, reviewing detected and tracked objects, searching for events, and assembling case evidence.

A Python video-processing pipeline exposes investigation workflows through FastAPI and SQLAlchemy, with detections and tracks persisted for search and case review. YOLOv8 performs detection; tracking, optional semantic search, optional Gemini/Ollama analysis, and evidence workflows are separate components.

PythonFastAPISQLAlchemyYOLOv8OpenCVNext.js
Kynovar research workbench screenshot from the repository's interactive-run artifactsSimulation / model evaluation

Kynovar

Research

Machine Learning Research / Simulation

A research system for controlled synthetic universes, simulation, dynamics-model training, and evaluation on observed trajectories.

Seeded simulators generate controlled dynamical systems and observable trajectories; data tooling prepares training and out-of-distribution splits for multiple dynamics models. Evaluation scripts record rollout and benchmark results, but the current repository explicitly marks research milestones and some results as work in progress.

PythonPyTorchNumPySciPySymPyLinear models

Additional Projects

NestAI secure code analysis dashboard

NestAI

Completed

Multi-Agent AI · Cybersecurity

An award-winning secure-code analysis CLI where adversarial Red and Blue AI agents evaluate software, a controller verifies findings, and the system generates a consolidated security report. Placed second out of 32 teams.

Built Red and Blue analysis agents plus a controller that correlates evidence, verifies findings, and produces a consolidated HTML report.

PythonMulti-Agent SystemsOpenAI APISecurity AnalysisCLI

Sentinel Log Ingestion

Private Enterprise Work

Non-native telemetry into Microsoft Sentinel

Log ingestion for non-native sources in Microsoft Sentinel, including Palo Alto firewall syslog via CEF.

Designed DCRs, custom tables, and schemas, then a syslog pipeline for Palo Alto logs with CEF forwarding and severity-based filtering.

Microsoft SentinelKQLAzureCEFLog Analytics

HiveU

Prototype

AI StudyMatch Platform

An AI study-partner matching platform that combines courses, availability, and study style to rank compatible peers, then gives each group a shared workspace for chat, files, and tasks.

Built a React client and FastAPI service that scores course overlap, availability, and study style, then supports group messaging, files, and tasks.

ReactFastAPISQLModelPythonSmart Matching

AI Recruiter Assistant

Prototype

Evidence-Grounded Candidate Copilot

An evidence-grounded recruiting copilot that evaluates candidate fit, verifies résumé claims against source material, and maps strengths to job requirements with transparent coverage scores.

Built retrieval signals across résumé, GitHub, and LinkedIn sources, then grounded evaluations and job-fit coverage in verifiable claims.

Next.jsPythonFastAPIPostgreSQLOpenAI API

Let’s build something useful.

Seeking backend, applied AI, and security engineering internships or new-grad roles in Atlanta.