Saurav Panigrahi

About

Fri, 01 May 2026 00:00:00 +0000

I am an engineer working across AI systems, research engineering, and AI safety.

My work has ranged from database autotuning with reinforcement learning to LLM post-training and alignment at Zoho Labs. I am especially interested in how model behavior changes under training, how systems should be evaluated, how tools affect reliability, and how failures surface in high-stakes settings.

I have also worked on AI safety research with Robert McCarthy at UCL and Lionel Levine and Jonathan Chang at Cornell, with a focus on self-preservation propensity, emergent misalignment, normative drift, and the side effects of character training.

Adaptive Sampling Networks

Fri, 01 May 2026 00:00:00 +0000

Co-authored with Navneel Singhal.

Adaptive Sampling Networks explore a simple question: can the decoding strategy of a language model be learned, instead of fixed by hand-tuned heuristics like temperature, top-k, or nucleus sampling?

Problem

Most LLM deployments treat decoding as a hyperparameter choice. The same sampling rule is applied across prompts, uncertainty regimes, and output distributions.

That is useful, but rigid. A sampler should be able to respond to the shape of the probability distribution it receives.

AI Safety Research Collaborations

Fri, 01 May 2026 00:00:00 +0000

Research collaborations with Robert McCarthy at UCL and Lionel Levine and Jonathan Chang at Cornell.

Focus

Self-preservation propensity in language models.
Emergent misalignment after narrow training interventions.
Normative drift due to emergent misalignment.
Side effects of character or persona training.

Questions

When a model resists shutdown or redirection, is the behavior instrumental or self-preservation-like?
How can self-preservation propensity be measured without relying only on surface-level refusal behavior?
Which training interventions create behavioral changes outside the intended target domain?
How do character or persona training procedures affect alignment-relevant behavior?

Artifacts

What This Connects To

This work sits at the intersection of model evaluation, behavioral generalization, and AI safety.

Emergent Misalignment

Fri, 01 May 2026 00:00:00 +0000

Selected references on emergent misalignment and broad behavioral shifts from narrow training signals.

Core

Emergent Misalignment: Narrow Finetuning Can Produce Broadly Misaligned LLMs
Introduces the central phenomenon: finetuning on a narrow harmful behavior can produce broader misaligned behavior outside the training domain.
Narrow Misalignment is Hard, Emergent Misalignment is Easy
Useful for thinking about why a broad misalignment direction may be a more stable and efficient solution than a narrow one.

Evaluation

Fri, 01 May 2026 00:00:00 +0000

Long-form references on benchmarks, measurement, and what evaluations actually test.

Biology And Scientific Evaluation

LAB-Bench
Benchmark for language models doing biology research tasks. Useful because it evaluates research-relevant behavior rather than only static factual recall.
FOMO26
Foundation model challenge for brain MRI, useful as a clinical-domain evaluation reference.

Methodology

Open Graph Benchmark
Standardized graph ML benchmark suite with datasets, loaders, and evaluators. Useful as a reference point for what benchmark infrastructure can look like.

Robotics And Sim-To-Real

RoboTwin
Dual-arm robot benchmark using generative digital twins for scalable task and data generation.

Medmarks

Fri, 01 May 2026 00:00:00 +0000

Medmarks is an open-source benchmark suite for evaluating medical capabilities in language models across a mix of verifiable and open-ended clinical tasks.

Focus

Medical LLM evaluation.
Verifiable and open-ended benchmark tasks.
LLM-as-judge evaluation for non-verifiable tasks.
Clinically relevant model capability tracking.

Artifacts

ML Systems

Fri, 01 May 2026 00:00:00 +0000

Long-form references on training, infrastructure, and implementation practice.

Training Systems

Frontier Model Training Methodologies
Survey of open frontier training recipes and implementation choices.
Scaling LLMs with JAX
Book-length treatment of distributed training practice.
Beyond Language Modeling: An Exploration of Multimodal Pretraining
From-scratch multimodal pretraining study with useful details on representation choices and scaling behavior.

Embeddings And Retrieval

How to Train the Best Embedding Model in the World
Detailed engineering writeup on embedding model training, label noise, verification, and dataset scale.

GPU Programming

CUDA Writeups by Tushar Gautam
Implementation-forward notes on CUDA kernels and optimization.

Now

Fri, 01 May 2026 00:00:00 +0000

Last updated: 2026-05-01

Focus

AI systems and research engineering.
Evaluation, tool use, and emergent misalignment.
The reliability problems that show up when AI systems operate in high-stakes domains.
Synthetic biology and bioinformatics, especially the bridge from digital design to physical validation.

Studying

ML systems engineering.
Synthetic biology foundations.
Bioinformatics algorithms.
Research workflows that produce useful technical judgment.

Looking To Meet

Researchers working on evaluation, tool use, and generalization.
Builders at the intersection of AI systems and biology.
Programs and communities that support independent AI safety research.

People

Fri, 01 May 2026 00:00:00 +0000

Curated people, labs, companies, and institutions.

This page should reveal taste, not networking ambition.

It should only include people or institutions whose work is directly useful for understanding AI safety, research engineering, ML systems, programmable biology, or industrial biology.

Plausible vs Faithful

Fri, 01 May 2026 00:00:00 +0000

Plausible reasoning sounds right.

Faithful reasoning preserves the structure of the thing being reasoned about.

That distinction matters because many failures do not look like nonsense. They look coherent. They explain themselves well. They use the right vocabulary. They produce an answer that could have been true.

The problem is that “could have been true” is a weak standard.

In writing, plausibility shows up as an argument that flows but hides a missing step. In research, it shows up as a result that has a clean story but rests on a proxy. In AI systems, it shows up as an answer that sounds grounded while drifting away from the actual process that produced it.

Programmable Biology

Fri, 01 May 2026 00:00:00 +0000

Long-form references on biological foundation models, structure prediction, and sequence modeling.

Genomic Foundation Models

Evo 2
Long-context genomic foundation model for sequence modeling and design.
HyenaDNA
Long-context sequence models at nucleotide resolution.

Structure Prediction

AlphaFold
Foundational protein structure prediction paper.
AlphaFold 3
Extends structure prediction toward biomolecular complexes and interactions.

Research Engineering

Fri, 01 May 2026 00:00:00 +0000

Selected references on research taste, engineering judgment, and doing useful technical work.

Research Taste

You and Your Research
Hamming’s classic essay on choosing important problems and organizing a life around serious work.
An Opinionated Guide to ML Research
Practical advice on developing taste and becoming effective in machine learning research.
Principles of Effective Research
A useful frame for research as a skill that can be deliberately improved.
Fast
Examples of ambitious work happening faster than conventional expectations.

Tool Use And Agents

Fri, 01 May 2026 00:00:00 +0000

Long-form references on tool use, agent environments, and reliability loops.

Agent Environments

Harness Engineering
Useful framing around agents as systems shaped by environments, specs, feedback, and reliability loops.
Code Mode
A concrete argument for exposing tools through code interfaces rather than forcing every step through chat-level tool calls.
Context Mode
A useful pattern for keeping agent context manageable when tools produce large or noisy outputs.

Training-Time Semantics

Agents Learn Their Runtime
Study of persistent versus reset Python interpreters in CodeAct-style training.

Engineering Practice

AI Gave Birth to the 100x Engineer
Long case study on compounding agent workflows with test harnesses and supporting tools.

Saurav Panigrahi

About

Adaptive Sampling Networks

Problem

AI Safety Research Collaborations

Focus

Questions

Artifacts

What This Connects To

Categories

Emergent Misalignment

Core

Evaluation

Biology And Scientific Evaluation

Methodology

Robotics And Sim-To-Real

Links

AI Safety

Medmarks

Focus

Artifacts

ML Systems

Training Systems

Embeddings And Retrieval

GPU Programming

Now

Focus

Studying

Looking To Meet

People

Plausible vs Faithful

Programmable Biology

Genomic Foundation Models

Structure Prediction

Research Engineering

Research Taste

Tool Use And Agents

Agent Environments

Training-Time Semantics

Engineering Practice