Hello!

I'm Chidansh Mehta

अहं नित्यं अध्येता अस्मि ||

Tinkerer

ML Engineer · Researcher

About Me

I build things that learn. Not because it's trending, but because there's something genuinely satisfying about watching a system improve at a problem you couldn't just hardcode your way out of.

My work spans the ML stack: from classical models and deep learning to reinforcement learning; applied across healthcare, environmental forecasting, and autonomous systems. But more than any specific domain, I care about whether something actually works outside the notebook.

I'm a try-athlete; not fast, not consistent, but always attempting something new. That approach carries into how I work: I'd rather try a method, fail fast, and learn than wait for perfect conditions.

Python PyTorch TensorFlow Scikit-learn Hugging Face Transformers SQL C++ AWS (SageMaker, Bedrock) Azure ML Docker YOLOv5 OpenCV NumPy Pandas PostgreSQL MongoDB Git Linux

Experience

Graduate Research Assistant

Syracuse University

June 2025 – Present
  • Conducting literature reviews on ML methodologies for time-series and radio signal analysis, synthesizing state-of-the-art approaches to guide research direction.
  • Implementing and benchmarking machine learning models on time-series and radio frequency signal data, evaluating performance across varying experimental conditions.
  • Developing preprocessing pipelines for raw signal data, including noise filtering, temporal feature extraction, and signal transformation for downstream model training.
  • Collaborating with faculty to design experiments, analyze results, and iteratively refine model approaches based on empirical findings.

Machine Learning Engineer

Cognifai

May 2023 – June 2024
  • Researched and evaluated transformer-based models for Automated ICSR Segmentation, classifying case reports into Valid, Potential, and Invalid categories.
  • Fine-tuned biomedical NER models (BioBERT, SciBERT, SciSpacy) on Hugging Face datasets to extract patients, drugs, and adverse events from medical literature.
  • Investigated prompt engineering with large language models for entity extraction, comparing results with supervised learning baselines.
  • Preprocessed large-scale PubMed datasets and conducted error analysis to evaluate real-world model performance in compliance-focused use cases.

Jr. Machine Learning Engineer

Omdena

July 2022 – March 2023
  • Collaborated with globally distributed teams on three projects across healthcare and environmental domains, delivering end-to-end ML solutions.
  • Trained classification models using RFID sensor data to detect urinary incontinence, improving caregiver response time in elderly care environments.
  • Built scalable pipelines for CAQI-based air quality forecasting from 300+ stations, integrating MSTL decomposition and spatial harmonization.
  • Applied advanced imputation methods, reverse geocoding, and temporal feature engineering to improve model robustness.

Jr. Data Science Intern

Zummit Infolabs

January 2022 – June 2022
  • Developed an audio-to-video synthesis pipeline using conditional GANs and 3D Morphable Models, achieving a 25% improvement in lip-sync accuracy.
  • Raised SSIM from 0.76 to 0.91 and reduced visual artifacts by 30% through temporal smoothing techniques.
  • Trained a PPO agent in CARLA simulator achieving 95%+ autonomy with intervention rates below 2/km across complex urban scenarios.
  • Improved training efficiency by 30% via VAE-based image encoding and upgraded object detection with YOLOv5 v6.1, increasing accuracy by 20%.

Projects

01

Automated ICSR Segmentation

Researched transformer-based models to classify Individual Case Safety Reports into Valid, Potential, and Invalid categories for pharmacovigilance compliance.

PyTorchHugging FaceBioBERTNLP
02

Biomedical Named Entity Recognition

Fine-tuned and benchmarked biomedical NER models on PubMed datasets to extract patients, drugs, and adverse events from medical literature using SciBERT and SciSpacy.

BioBERTSciBERTSciSpacyHugging Face
03

Air Quality Forecasting Pipeline

Built scalable pipelines for CAQI-based air quality forecasting using pollutant and weather time-series from 300+ stations, integrating MSTL decomposition and spatial harmonization.

XGBoostLightGBMGeoPandasPython
04

Talking Face Synthesis

Developed an audio-to-video synthesis pipeline using conditional GANs and 3D Morphable Models to generate photorealistic talking-face animations from raw speech. Achieved 25% improvement in lip-sync accuracy and raised SSIM from 0.76 to 0.91.

PyTorchOpenCVDlibLibrosaGANs
05

Autonomous Driving Agent (CARLA)

Trained a PPO reinforcement learning agent in a Gym-wrapped CARLA simulator, achieving 95%+ autonomy with intervention rates below 2/km across complex urban driving scenarios.

PPOStable-BaselinesYOLOv5CARLAVAE
06

Spinal Rehabilitation ML System

Engineered predictive models from 3D spinal movement data to provide personalized and gamified rehabilitation plans for back pain relief using signal processing and motion analytics.

Signal ProcessingScikit-learnPython

Education

M.S. in Computer Science

Syracuse University

August 2024 – May 2026
  • Relevant coursework: Algorithms, Machine Learning, Operating Systems, Advanced Computer Architecture, Introduction to AI

B.E. in Computer Engineering

Gujarat Technological University

September 2018 – June 2022
  • Relevant coursework: Data Structures, Algorithms, Computer Architecture, Operating Systems, Computer Networks, Cloud Computing, Software Engineering