aher.dev
Jaideep Aher

Seeking Summer 2026 internships in AI Product, ML, and Data

Durham, NC · Open to relocate

Jaideep Aher

I turn AI prototypes into shipped products

AI Product Builder | ML Systems | Data

Duke AI master's student building ML systems, data products, and 0→1 AI tools. Previously worked on L3 ADAS programs at Magna and founded AI products in fleet safety and resale automation.

3 patents3 hackathon winsEx-Magna ADASDuke AI MEng

Projects

Selected work

Finished products in market across commerce, agri-tech, fleet safety, health, and applied AI, plus Duke coursework and research demos. Built to be used, not only presented.

  • Course Co-Pilot

    Multi-agent pipeline for transfer credit evaluation: a Parser Agent extracts courses via GPT-4o vision, Research Agents autonomously search university catalogs in real time, and an Orchestrator re-triggers research on low-confidence results before an Evaluation Agent finalizes scores.

  • ThriftSpark

    Run your entire thrift store from one platform: POS, inventory, multi-channel selling, analytics, and shipping in one place. AI-powered listings in about 30 seconds.

  • SafeType

    AI-powered parental monitoring for Android: custom DistilBERT over 10 safety categories, real-time alerts across WhatsApp, Instagram, Snapchat, and more, with transparent, child-visible monitoring (Duke deep learning project).

  • VoiceClone AAC

    iOS app that gives ALS patients their voice back: cloned from a short sample, it speaks any phrase they type.

  • Agrofix

    Scalable agri-tech solutions for growth: autonomous agriculture with smart spraying that cuts chemical waste and lifts yields for modern Indian farms.

  • Fleet Kawach

    AI-powered fleet safety: prevent accidents, save lives. Integrated platform for drunk driving prevention and real-time AI safety coaching for fleets in India and beyond.

  • Driver Monitoring System

    Occupant-state stack: dense facial mesh with refined iris loci, EAR/MAR analogs, PnP head pose with geometric fallback, gaze_region lattice for visual escape vs. forward field. Collapsed metrics only (no raw frame egress); versioned telemetry envelope into FastAPI; RollingSignalBuffer windows, parallel DistractionMonitor with orthogonal cooldowns vs. fatigue, semantic synthesis head with SHA-keyed deduplication. SQLAlchemy 2.0 async → PostgreSQL (TLS); trips + append-only alert ledger for replay. Dockerized ASGI; canonical dms.aher.dev.

  • Hundekari

    Farmer Android app on Google Play with 2K+ downloads. Together with the farmer web platform, the ecosystem is visited by ~4K unique farmers each month.

  • HHG Farmer Portal

    Digital receipts for sold produce, Mandi market rates with 7-day history, and a locality-based vendor marketplace. ERP-synced receipts from middlemen, trend analysis to time sales, verified local buyers and sellers, and yield insights from Mandi performance.

  • Coral Reef Monitor

    AI coral reef health monitoring with computer vision.

  • Duke Reuse 360

    Skip the line, save the planet. DukeReuse 360 is Duke University's AI-powered smart reusable dining container system. Pre-book your container, skip the wait, and track your environmental impact in real time.

  • Kitchen Verdict

    Fine-tuned OpenAI model with a Gordon Ramsay–style chef persona baked into the weights; a side-by-side compare mode lets you see fine-tuned vs. base model outputs in real time, with an adjustable temperature parameter to control output sharpness.

  • KITTI LiDAR visualizer

    Compact browser viewer for KITTI-style Velodyne scans: distance or intensity coloring, optional calib and label overlays, Three.js; handy for ADAS / LiDAR format demos, not a shipped product.

  • MediAlert

    Advanced custom trained deep learning models that detect pneumonia in chest X-rays in seconds, not hours. Empowering doctors with instant, accurate clinical insights.

GitHub

Contributions

Public commit activity over the last year. @jaideep-aher

jaideep-aher on GitHub: year of contributions

Rendered by githubchart from public contribution data; it may differ slightly from GitHub's own chart.

Contact

Let's connect

Open to Summer 2026 roles where I can ship AI products end to end, whether in engineering, data science, or product. Reach out by email or LinkedIn. You can also book a time that works for both of us below, or use aher.dev/meet as a standalone link.