Research Engineer
Atul Saxena
I build the environments models are trained in, and the tools that show what they learned.
- RL Systems
- Interpretability
- ML Infrastructure
Research Engineer
Atul Saxena
I build the environments models are trained in, and the tools that show what they learned.
Research infrastructure and tools. The ones marked with a package are installable today — not demos.
Photorealistic drone simulation for vision-language-action research
A pip-installable drone environment that combines 6-DOF quadrotor dynamics with photorealistic Gaussian-Splatting rendering, and the data tooling to train and evaluate VLA policies on it.
Watch a language model think, token by token
Hooks GPT-2’s residual stream at layer 8, runs each token’s hidden state through a sparse autoencoder, and streams the top-K firing features into a live 3D force graph in the browser.
Tiny installable behaviour patches for frozen language models
A portable file format and runtime for reversible activation-space interventions. No fine-tuning, no weight modification, no prompt injection, no hosted service.
Mathematical search you can inspect
An evidence-aware neuro-symbolic mathematics platform in Rust. It searches a stable registry of rules, records every transformation it applies, and returns the result with a precise verification status.
Calibrated decisions in milliseconds, self-hosted
An open decision model for agents and pipelines. Give it a state and typed questions; get back calibrated probabilities, an abstain signal and an uncertainty measure. It never writes text.
Synthesizing verifiable reward signals outside maths and code
Catching silent corruption in distributed GPU training
Will this model fit, and how fast will it run?
Stop API keys before they reach git
Allocator-level heap telemetry for Zig
Agentic OSINT geolocation on 4 GB of VRAM