Hi, I'm Pratham
Data Scientist with 2+ years of experience shipping production ML systems, NLP pipelines, and computer vision solutions on cloud-native infrastructure (AWS, EKS, Docker). I translate business problems into deployed AI products — and mentor the next generation of engineers along the way.
Featured Projects
View all →AI-Powered Meeting Analyser
NLP-powered meeting intelligence platform that automates transcription, summarization, and action item extraction with an event-driven pipeline, saving 3+ hours weekly per user.
No-Code ML Platform
Enterprise-grade no-code ML platform with time series forecasting module, reducing baseline model development time by 80% for forecasting workflows.
SQL Agent
Autonomous text-to-SQL agent that converts natural language into validated, executable queries over a live PostgreSQL warehouse — letting non-technical teams self-serve reports in minutes instead of waiting hours on analysts.
Latest Posts
Read all →Kolibri Post-Training from Scratch: Effort Labels, Format Rewards and a Gradient-Norm Trust Region
Building Aleph Alpha's Kolibri post-training recipe from the ground up: soft reasoning-effort labels drawn from log-length percentiles, attention-cost ordering for packed SFT batches, segment-aware group advantages, the p-norm format multiplier, the B-TV trust region, a closed-form per-token bound on the logit gradient, targeted on-policy self-distillation, the three-bucket curriculum, and the Merlin–Arthur game against hallucination. Every piece is derived on one German baker problem and a group of four rollouts.
Kolibri Pre-Training from Scratch: UniBPE, Hybrid Attention and Exact Quantile Balancing
Building Aleph Alpha's Kolibri base model from the ground up: the Unigram-loss merge score of UniBPE and why it is pointwise mutual information in disguise, the KV-cache arithmetic of 40 sliding-window and 10 NoPE blocks, sigmoid MoE routing with Exact Quantile Balancing and Load-Error Injection, the 1 − K/E signature of a router the bias has overridden, and the cumulative-learning-rate law behind hyperparameter transfer. Every step is derived on one German sentence and four tokens.
MiMo-V2.6: Scaling Reinforcement Learning from Scratch
Building Xiaomi's MiMo-V2.6 RL recipe from the ground up — the importance-sampled GRPO objective, group-relative advantages and dynamic sampling, hack zeroing, Groupwise Reward Synthesis, Groupwise Advantage Redistribution, the group-relative length penalty, mass-conserving segment penalties, top-p candidate replay, router freezing, and the Little's-law Sample Mixer — all derived step by step on one four-rollout coding group.
Let's work together
I'm always open to discussing new projects, creative ideas, or opportunities to be part of your vision.
Get in Touch