Supriyo Banerjea profile photo 1 on Tal

Supriyo Banerjea

M.Tech (Research) Al @ IITH’27 • AIR 336, GATE DA 2024

Kandi

2+ years of work experience

Indian Institute of Technology Hyderabad
Jul 20242 yrs 2 mosPresent

Hyderabad, Telangana, India

MACROMILL, INC.
May 20263 mosJul 2026

Shinagawa-ku, Tokyo, Japan

· Built a probabilistic 7-day survey response forecasting system to optimize panel capacity planning and real-time field operations. · Engineered a Hierarchical Bayesian Survival model with natural depletion using PyMC and JAX, blending ML propensity scores with real-time Laplace posterior updates. · Implemented a Monte Carlo engine simulating 10,000 trajectories, dynamically blending CDFs to mitigate early-data noise. · Validated model reliability on a chronological 500-project hold-out set via a leakage-safe PySpark pipeline on Databricks, achieving 93.5% CI coverage with 24 hours of observed data.

Supriyo’s AI Stack

antigravity
antigravity
cursor
cursor
opencode
opencode

Education

Indian institute of technology Hyderabad

Master of Technology - MTech, Artificial Intelligence

Indian institute of technology Hyderabad

20243 yrs2027
Heritage Institute of Technology

Bachelor of Technology - BTech, Computer Science and Engineering

Heritage Institute of Technology

20204 yrs2024

Join Supriyo and 10k+ other professionals on

Tal Boss

Tal is the app where bosses meet engineers & builders directly, without middlemen

Download Tal Boss

Supriyo Banerjea on Tal

Headline

M.Tech (Research) Al @ IITH’27 • AIR 336, GATE DA 2024

Location

Kandi

AI stack

antigravity

cursor

opencode

Work experience

Research Assistant at Indian Institute of Technology Hyderabad, Jul 2024 - Present · 2 yrs 2 mos, Hyderabad, Telangana, India

Data Scientist at MACROMILL, INC., May 2026 - Jul 2026 · 3 mos, Shinagawa-ku, Tokyo, Japan, · Built a probabilistic 7-day survey response forecasting system to optimize panel capacity planning and real-time field operations. · Engineered a Hierarchical Bayesian Survival model with natural depletion using PyMC and JAX, blending ML propensity scores with real-time Laplace posterior updates. · Implemented a Monte Carlo engine simulating 10,000 trajectories, dynamically blending CDFs to mitigate early-data noise. · Validated model reliability on a chronological 500-project hold-out set via a leakage-safe PySpark pipeline on Databricks, achieving 93.5% CI coverage with 24 hours of observed data.

Education

Master of Technology - MTech, Artificial Intelligence, Indian institute of technology Hyderabad

Bachelor of Technology - BTech, Computer Science and Engineering, Heritage Institute of Technology

Public links

github: https://github.com/sycoraxx

Supriyo Banerjea - Research Assistant in Kandi | Tal