B.Tech Chemical Engineering, 2nd year — NIT Calicut. Reactors, models, and a fair bit of Python. Portfolio
Edition 2026 ✱

Allen

A chemical engineering student who likes modeling reactors and writing code for them. Currently interning at Sud-Chemie India.

Scroll

Chemical EngineeringProcess ModelingMachine LearningQuant + CarbonChemical EngineeringProcess ModelingMachine LearningQuant + Carbon

(01)

About

I'm a second-year Chemical Engineering student at NIT Calicut, based in Cochin. I mostly use code to poke at reaction-engineering problems — lately that's been hybrid ML and first-principles reactor models, plus a couple of quant side-projects I'm still figuring out.

Fig.01 — AllenCochin, 2026

Allen Binoy

Based inCochin, India
Born31.08.2005
StudyingChem. Eng · NITC
CurrentlyIntern · Sud-Chemie
FocusProcess · ML · Quant

(02)

Education

01 B.Tech Chemical Eng. National Institute of Technology Calicut — currently in 2nd year. Reaction engineering, thermodynamics, transport phenomena and process modeling. 2024 —
2028 ↗

(03)

Experience

01 Team Lead Heeszo Platforms — leading the engineering for Heeztag, a hybrid mobile app that combines a high-traffic photo-sharing network with an e-commerce marketplace. I oversee system architecture, code quality and team execution to build a scalable, high-performance platform. Full-time, remote. Mar 2026
Present ↗

02 Engineering Intern Sud-Chemie India Pvt. Ltd., Cochin — summer internship at a specialty-chemicals & catalyst manufacturer. May 9 – July 17, 2026. May —
Jul 2026 ↗

(04)

Selected Work

01 Predictive Maintenance DashboardXGBoostFastAPIChart.js A full-stack predictive-maintenance app for a shell-and-tube heat exchanger — a physics simulator (effectiveness-NTU + Kern-Seaton fouling) feeds XGBoost/RandomForest models that estimate State of Health and Remaining Useful Life, streamed live to a Chart.js dashboard with a maintenance verdict. Jun 18
2026 ↗ 02 Deep RL Energy ControllerNumPyDQNFastAPI A Double-DQN agent that trims reboiler energy on a distillation column while holding product purity above spec — roughly 20% less energy than a fixed-setpoint controller. Built from scratch in pure NumPy with a custom Gym-style environment and a live dashboard. Jun 19
2026 ↗ 03 Distillation Soft Sensorscikit-learnSVRFastAPI A machine-learning soft sensor that infers ethanol/water distillate purity in real time from tray temperatures, pressure and reflux — trained on a first-principles vapour-liquid-equilibrium column model and served through a FastAPI backend with a live dashboard. Jun 20
2026 ↗ 04 CSTR Hybrid Neural Networkscikit-learnSciPy A neural-network surrogate for a continuous stirred-tank reactor — trained on first-principles ODE simulations to predict steady-state concentrations orders of magnitude faster, for real-time model-predictive control. Handles irreversible, reversible & parallel kinetics. May 6
2026 ↗