saksham@dev:~$
saksham.jpg
Based
Dehradun, UK
CGPA
8.7 / 10
Engage
Freelance / Full-time
Stack
Python · FastAPI · PyTorch
Status
Open to work
stack.json
Technical Skills
Languages
PythonJava
SQLC
Backend & APIs
FastAPIREST APIs
LLM / GenAI
Gemini APIStructured Outputs
RAG
AI / ML
PyTorchscikit-learn
NumPypandas
CNNGRU/LSTMKAN
Databases & Infra
MySQLMongoDB
PostgreSQLVector Search
GitORM
Desktop / GUI
Java Swing
./what-i-do
What I Do
Full-Stack AI
End-to-end AI apps: FastAPI backends, LLM/RAG pipelines, and clean, usable frontends.
Deep Learning
Designing, training, and benchmarking models from scratch — CNNs, GRU/LSTM, and KAN.
Backend & APIs
FastAPI, REST, relational & vector databases — built with production in mind.
LLM / GenAI
Gemini API, structured outputs, and RAG grounding to cut hallucination in real apps.
./projects
Selected Work
Full-Stack AI
Live
JobGenie
A full-stack AI copilot with a FastAPI backend and Gemini API for resume parsing, ATS-score simulation, and job-description matching. A RAG pipeline grounds LLM outputs to reduce hallucination in match scoring, with an analytics dashboard, multi-resume A/B testing, and interview tracking.
Python · FastAPI · Gemini API · RAG · JavaScript
● AI-powered job-search assistant
jobgenie.tech ↗
Deep Learning
5th · CSA Hackathon 3.0
PhoniX
A real-time dialer that detects voice-phishing (vishing) attacks by classifying live audio streams. A deep-learning pipeline distinguishes authentic human speech from deepfake-generated audio using spectrogram-based feature extraction, evaluated via precision, recall, and F1.
Python · PyTorch · Audio Processing · Deep Learning
● 89.68% detection accuracy
Deep Learning
Research
Session-Based Recommender
Next-item recommendation on the YooChoose clickstream dataset (50K sessions, 128K+ samples). Implemented and benchmarked three architectures from scratch — GRU baseline, pure recurrent KAN, and a hybrid GRU+KAN with learnable B-spline activations. Trained with AdamW and cosine LR scheduling.
Python · PyTorch · GRU + KAN · Deep Learning
● 58.08% HR@10 · 39.42% NDCG@10
Full-Stack
Open Source
Launch Cadence
A rocket-launch management and booking system — scheduling, management, and booking workflows for launches, built end-to-end.
Full-Stack · Booking System
Desktop App
Java · Swing
Library Management System
A desktop GUI for managing books, members, and issue/return workflows, backed by a relational database.
Java · Swing · SQL · Desktop GUI
git log --experience
Experience & Leadership
commit c5a0891
President · Registrations Head
Cloud Security Alliance (CSA), UPES
ongoing
2025 — 2026
- Lead the UPES CSA chapter, directing cross-functional technical, event-management, and communications teams on community-wide cloud-security initiatives
- Managed end-to-end registration workflows for a 400+ participant event as Registrations Head, coordinating onboarding across teams to cut process bottlenecks
- Drove outreach initiatives that grew measurable community engagement around cloud-security awareness
commit 3f7b214
Social Intern
Gram Rajya Vikas Evam Prashikshan Sansthan
Jun — Jul 2024
- Assisted legal authorities in more than 5 rescue and rehabilitation operations related to child-labour cases
- Supported more than 20 interventions under child protection and POCSO regulations
// Field Coordination · Documentation · Community Engagement · Teamwork
./education
Education
University of Petroleum & Energy Studies (UPES), Dehradun
B.Tech, Computer Science & Engineering (AI & ML)
Relevant coursework: Machine Learning · Pattern & Visual Recognition · Deep Learning · Data Structures & Algorithms · Database Management Systems · Operating Systems
8.69
CGPA / 10.00
cat about.md
About
I don't like leaving ideas as ideas. If I can imagine it, I want to build it.
I'm a final-year Computer Science (AI & ML) student at UPES who enjoys turning concepts into real products. From training machine learning models and building FastAPI backends to integrating LLMs with RAG, I love working across the entire stack. Whether it's JobGenie or my latest experiment, I'm always chasing the next problem worth solving.
Outside of building, you'll usually find me solving DSA, exploring new technologies, or figuring out how to make software just a little smarter.