Work

Jan 2026 – Feb 2026 Pune

Backend Developer

KissanKonnect
  • Built and deployed the official production website for a high-growth agritech company — live and serving real users from day one.
  • Engineered full backend architecture: Express.js, Supabase auth, role-based access control for multi-stakeholder use, and Razorpay payment gateway.
  • Built RESTful APIs powering the Next.js frontend; delivered production deployment within 4 weeks through MLSC.
Next.jsExpress.jsSupabaseRazorpay
Aug 2025 – Present Pune

AI/ML Coordinator

Microsoft Learn Student Club, VIT Pune
  • Spearhead the AI/ML vertical — organizing 10+ workshops and speaker sessions on Agentic AI, NLP, and computer vision for 100+ student members.
  • Source and coordinate real industry projects for the club, including the KissanKonnect production deployment.
  • Mentor junior developers on model training, deployment best practices, and responsible AI use across cohorts.
Agentic AINLPMentorshipLeadership
Jun 2025 – Jul 2025 Pune

Artificial Intelligence Engineer

Aria Technologies
  • Sole engineer on a production-grade NL-to-SQL chatbot converting natural language queries to optimized database operations on a proprietary product catalog.
  • Designed a multi-layer security architecture: threat modeling, prompt injection safeguards, and data anonymization to prevent sensitive data exposure.
  • Delivered end-to-end — LLM fine-tuning, semantic schema mapping, contextual few-shot learning, and API integration — deployed to internal stakeholders in 6 weeks.
PythonLLMsNL-to-SQLREST APIs
Sep 2024 – Jun 2025 Pune

AI/ML Developer

Google Developer Group
  • Built and trained ML models for predictive analytics, NLP, and computer vision; optimized for performance and scalability with TensorFlow, PyTorch, and Google Cloud ML tools.
  • Organized knowledge-sharing sessions and mentored 20+ developers on transformer architectures, model optimization, and responsible AI practices.
TensorFlowPyTorchGoogle Cloud ML
Sep 2024 – Jun 2025 Pune

Computer Vision Head

Team Griffin India AIR 1 · SAE 2025
  • Architected the full CV stack for an autonomous aeromodelling platform: YOLOv8 detection, multi-object tracking, and ROI pipelines optimized for NVIDIA Jetson Nano edge inference.
  • Led a 5+ member engineering team through ideation, model training, and real-time deployment under stringent latency constraints.
  • Directly contributed to winning AIR 1 (Overall) and AIR 1 (Design Report) at the SAE Aeromodelling Challenge 2025, Chennai.
YOLOv8OpenCVJetson NanoPython
Jun 2024 – Sep 2024 Mhaswad

Lab Assistant — C Programming

Seed Infotech Ltd
  • First professional role — assisted 30+ students in C programming lab sessions, helping debug programs and master core concepts.
  • Prepared and organized lab sessions; gained early exposure to mentoring in a technical setting.
CTeachingMentorship
PUBLICATION

Adaptive TabNet–β-VAE Fusion Ensemble Closes Uncertainty Gap In Fraud Detection

First Author  ·  ICAIN 2025  ·  Scopus & Springer Indexed (in press)

Projects

01

SGEE + ACA — NLP Emotion & Persuasion Pipeline

VIEW →

Multi-stage deep learning pipeline for fine-grained emotion analysis and persuasion detection. Custom SGEE layer fuses RoBERTa embeddings with NRC-VAD lexicon features into a 512-dim emotion space. 6 specialized prediction heads across 28 emotion classes, VAD regression, and persuasion tactics.

PyTorchRoBERTaHugging Facescikit-learn
02

EdgeCAAI-Net — Artist-Invariant Genre Classifier

VIEW →

Lightweight Conformer-Lite architecture for music genre classification with artist-disjoint evaluation. Exposes artist leakage in standard benchmarks (18.8% F1 drop) using gradient reversal for artist-invariant representations. Multi-exit early stopping for edge deployment.

PyTorchConformerGRLMel Spectrograms
03

Unified Multimodal Emotion Analysis Platform

PRIVATE

End-to-end emotion detection across text, image, audio, and video via a unified REST API. Custom CNN → BiLSTM → Transformer → Attention model (UltimateNet) for audio. 7-stage video pipeline with Whisper transcription, EasyOCR, and Cerebras LLM cross-modal fusion.

PyTorchFlaskOpenCVWhisperCerebras LLM
04

Fraud Detection Suite

PRIVATE

1M synthetic transactions with realistic device/account behavior simulation. TabNet–β-VAE ensemble for anomaly detection. Published as first-author research paper at ICAIN 2025 (Scopus & Springer indexed).

PythonTabNetβ-VAEscikit-learn
05

WhatsApp Clone — Real-Time Chat

VIEW →

Multi-threaded client-server chat app built from scratch in Java. Custom TCP/IP socket layer, thread-per-client architecture with ConcurrentHashMap, normalized MySQL schema, MVC pattern, and custom Swing UI with message bubbles, presence detection, and read receipts.

JavaMySQLJDBCSocket ProgrammingSwing
06

LocateVision — Hybrid Image Classifier

VIEW →

Hybrid ResNet-50 + Swin Transformer classifier with a custom Gated Attention mechanism. Trained with AdamW + CosineAnnealingWarmRestarts. Served via Flask REST API with a full web UI — drag-and-drop upload, prediction history, dark mode, and user auth.

PyTorchResNet-50Swin TransformerFlaskAlbumentations
07

Autonomous Drone CV Stack — Team Griffin

PRIVATE

YOLOv8 detection + tracking + ROI pipeline for autonomous drone missions. Optimized for edge deployment on NVIDIA Jetson Nano under real-time latency constraints. Core system behind Team Griffin's AIR 1 performance at SAE Aeromodelling Challenge 2025.

YOLOv8OpenCVJetson NanoPython
08

Banking Management System

PRIVATE

15,000+ lines of C across 15+ modules. Full-featured banking platform with multi-session concurrency (Win32 threads), flat-file persistence, live stock market simulation (multi-factor price model), quantitative analytics, 4-heuristic fraud detection engine, 2FA, and an admin panel.

CWin32 APIMultithreadingFlat-File I/O
09

KissanKonnect — Industry Production Website

VIEW →

Official production website for a high-growth agritech company — live and serving real users. Full backend: Express.js, Supabase auth + database, role-based access control, and Razorpay payment gateway. Delivered in 4 weeks via MLSC industry project.

Next.jsExpress.jsSupabaseRazorpay

Skills

LANGUAGES
PythonCC++ JavaJavaScriptSQLNoSQL
DEEP LEARNING
PyTorchTensorFlowHugging Face CNNsTransformersRNNs/LSTMs GANsAutoencodersAttention Multi-Task LearningContrastive Learning
NLP / LLMs
RoBERTaLLM Fine-tuningPrompt Engineering NL-to-SQLWhisperEasyOCR Agentic AINLTKVADER
CV / ML LIBRARIES
YOLOv8OpenCVscikit-learn torchvisiontimmAlbumentations LibrosaNumPyPandas
WEB / BACKEND
Next.jsExpress.jsFlask SupabaseRazorpayREST APIs MySQLJDBCHTML/CSS/JS
INFRASTRUCTURE
GitLinuxDocker FFmpegJetson NanoWin32 APIPydub
EDUCATION
Vishwakarma Institute of Technology, Pune
B.Tech Computer Science — AI & Machine Learning  ·  2024 – 2028
CGPA 9.05 / 10  ·  First Year

Outside the Terminal

What I do when I'm not building things — and why it matters.

01

Powerlifting

Discipline over motivation. The barbell doesn't care about your mood — you either lift it or you don't. Progressive overload taught me compounding before any textbook did.

02

Motorcycling

Forced mindfulness — real-time systems thinking at 80 km/h. You can't be distracted on a bike. Some of my best ideas come mid-ride, and some of my clearest decisions do too.

03

Swimming

Technique over brute force. Trying harder doesn't mean going faster — reduce drag before you increase power. That lesson transferred directly to how I approach engineering problems.

04

Badminton

Fast, reactive, and deeply strategic. Doubles sharpened my teamwork instincts — anticipate what your teammate needs, cover their gaps, and trust them to cover yours.

05

Tabla & Piano

Both instruments taught me layered thinking — rhythm underneath melody, structure underneath expression. Same as designing a good system: every component has its role.

06

Beatboxing

Constraints breed creativity. One instrument, one challenge: make it sound like a full production. Same principle that built a 15,000-line banking system in C with no external libraries.

Résumé

One page. Every signal that matters.

Contact

Always open to interesting builds, research, and conversations.

© 2026 Ameya Borkar · Designed & built from scratch