Hi

I'm

Siddhant Rajhans.

I'm a Machine Learning Engineer specializing in production AI systems, agentic architectures, and multimodal RAG. My passion is building intelligent systems that process 400M+ records/month and push the boundaries of what's possible with frontier models.

→ Say Hi
Siddhant Rajhans

Explore My Latest Projects

CortexLab Architecture

CortexLab: Brain Encoding Toolkit

Multimodal fMRI brain encoding toolkit built on Meta's TRIBE v2. GPU voxelwise ridge (Triton), causal modality lesion analysis, brain-alignment benchmarking, 3D brain viewer, live inference from webcam/screen/video. 143 tests, 4 contributors, published on PyPI and HuggingFace.

PyTorchTritonfMRI3D VizLive Inference
Bloomberg Sentiment x FinBERT Architecture

Bloomberg Sentiment × FinBERT

Empirical test of whether Bloomberg's NEWS_SENTIMENT_DAILY_AVG can be replicated with open-source FinBERT on a 30-stock S&P 100 universe (2018-2026, 62.8K stock-day obs). Multi-horizon IC, long-short quintile backtest, Fama-French 3 with Newey-West HAC. Replication fails (ρ = −0.26) — and that's the point.

Bloomberg APIFinBERTstatsmodelsFama-FrenchNewey-West
DreamStudio Architecture

DreamStudio: AI Cinematic Story Director

AI-powered cinematic story director. Speak naturally, point your camera, and watch scenes materialize as images, video, and music in real-time. Full-stack app with web, mobile, and Python backend.

Gemini LiveImagen 4Veo 3.1Lyria 2FlutterPython
Data Pipeline Architecture

Messaging-Based Data Pipeline

End-to-end streaming pipeline processing 400M+ records/month with 65s latency (p95). Kappa architecture with event-time processing, watermarking, and fault tolerance built for bursty event ingestion.

KafkaPySparkDockerAirflowHiveNiFi
Multimodal RAG Architecture

Multimodal RAG for Scientific Literature

Textual + multimodal RAG system for multi-turn conversational queries over 500K+ papers. Agent evaluation metrics for grounding across 2.3M figures, 890K tables, 410K equations. 94% recall at sub-100ms latency.

LangGraphSciBERTCLIPTAPASFAISSOCR
Bike Lane Sentinel Architecture

Bike Lane Sentinel

Computer vision system that monitors illegal vehicle encroachment into NYC bike lanes and automatically alerts NYC DOT. Real-time detection with interactive dashboard and evidence capture.

Computer VisionTypeScriptReactNode.jsReal-time

Built Production Systems & Shipped Real Products.

2023 - 2024

Machine Learning Engineer

Seed-Stage Health-Tech Startup (NDA)

  • Designed and shipped streaming ML pipelines on AWS (Kafka, Airflow, Docker) with the data engineering team, processing 400M+ healthcare records/month at <65s p95 latency.
  • Developed transformer-based NLP and multimodal retrieval pipelines for clinical notes, lab reports, and diagnostic documents, improving retrieval accuracy by 30%.
  • Led the retrieval and prompt-orchestration workstream of an agentic AI system powering adaptive, personalized clinical-facing interactions.
  • Built model monitoring and data validation checks tracking data drift, prediction quality, and data integrity across clinical ML workflows.
AWSKafkaAirflowDockerNLPAgentic AI
2023

Software Development Engineer Intern

Seed-Stage Health-Tech Startup (NDA)

  • Contributed to developing backend services and APIs supporting ML-driven features, helping reduce average request latency by 40% by moving synchronous flows to async processing.
  • Helped improve ML model serving pipelines with caching and request batching, making clinical-facing features more responsive.
  • Assisted senior engineers in strengthening API security: input validation, authentication middleware, and audit logging for sensitive healthcare data.
ML ServingAPIsBackend OptimizationSecurity

Technologies and Tools I Work With

Core ML & Deep Learning

PyTorchTensorFlowScikit-learnXGBoostLightGBMTransformersHugging Face

LLM, Agents & Evaluation

LangGraphRAGAgentic OrchestrationPrompt EngineeringEvaluation FrameworksBenchmark DesignGuardrails

Data & Infrastructure

PythonSQLApache SparkKafkaRedisFAISSPostgres/PgvectorS3OpenSearch

MLOps & Cloud

AWSSageMakerEKSEMRDockerAirflowMLflowW&BGrafana

Research Publications

IEEE ICCE 2025

CNN-Based Detection Mechanism for Deepfake Image

A Mishra, S Rajhans, BB Gupta, KT Chui

Springer, ICRTC 2025

The Evolving Landscape of Cloud Computing: AI Integration, Threats, Challenges and Security Concerns

P Singh, S Rajhans

Springer, ICSPN 2023

Machine Learning and AI in Cybersecurity: Insights and Solutions

S Rajhans, A Mishra

Education & Credentials

2025–2026

MS in Machine Learning

Stevens Institute of Technology

Hoboken, NJ
 

B.Tech in Computer Science & Engineering

Swami Rama Himalayan University

Dehradun, India
Training and Fine-tuning LLMs, W&B Machine Learning, Stanford / Coursera Generative AI with Diffusion Models, NVIDIA Cloud Computing Diploma, IBM