# Kshitij User

Kshitij User is a remote worker on Himalayas, Founder & Research Engineer at TENS LAB, based in India.

Source: https://himalayas.app/@kshitijpal

```text
Founder & Research Engineer at TENS LAB building CONTEX, a token compiler for deterministic LLM inference.
```

| Detail | Value |
| --- | --- |
| Availability | Open to opportunities |
| Location | India |
| Authorized to work in | India |
| Salary expectations | 16600k+ INR |
| Job categories | Research Engineer, Founder |
| Skills | Compiler Development, System Architecture, Tokenization, LLM, Hybrid Model Merging, Knowledge Distillation, Agentic Engineering, AI ML, Relational Databases, Redis, DBMS, Content Delivery Networks (CDN), Cloud Computing, IoT, Fullstack Development |
| Website | https://tenslab.dev |
| LinkedIn | https://linkedin.com/in/kshitijpalsinghtomar |
| GitHub | https://github.com/tens-lab |

## About

```text
At TENS LAB, I’m building deterministic context representation and token compilation systems for language model inference. CONTEX uses a canonical intermediate representation to support KV prefix-cache reuse and deterministic inference across LLM runtimes.

I’m also developing CONTEXDB, an inference-native context storage engine designed around prefix-cache locality. My research benchmarks latency, compute, and token cost from prompt syntax overhead in frontier LLM architectures.

I’m still early in my learning, and I’m intentional about understanding how technology and systems work. I explore ideas through self-learning, research, and small experiments, with interests in system design and how technology scales in real environments.
```

## Experience

### Founder & Research Engineer (current)

TENS LAB. Jul 2025 - Present

```text
Building deterministic context representation and token compilation systems for language model inference at TENS LAB.
Research & Systems:
• CONTEX — Token compiler implementing a canonical intermediate representation (IR) to maximize KV prefix-cache reuse, eliminate redundant prompt tokens, and ensure deterministic inference across LLM runtimes.
• CONTEXDB — Inference-native context storage engine
```

### Independent Learner / Project Exploration (current)

Self-employed. Aug 2023 - Present

```text
Exploring technology, digital platforms, and product ideas through self-learning, research, and small experiments.

Interests include system design, platform thinking, and how technology scales in real environments
```

## Education

### Parul University

Bachelor of Technology, Computer Science. 2024 - 2028

## Tech stack

- React Native
- Node.js
- Next.js
- Redis

To contact Kshitij, message them from this profile with a Himalayas employer account: [Hire on Himalayas](https://himalayas.app/hire).

## People also viewed

- [Devansh Vora](https://himalayas.app/@devanshvora): Jr. AI Research Fellow, Smart Energy Learning Center Dhirubhai Ambani University
- [Krishna Yadav](https://himalayas.app/@krishnayadav4): AI Engineer (LangGraph), Elvz.ai
- [Vinay Soni](https://himalayas.app/@vinaysoni): Independent AI Research Engineer, Independent
- [Yash User](https://himalayas.app/@yashuser20): Solo Developer, Stepwise
- [Nirmal V](https://himalayas.app/@nirmalv): Founding Software & AI Engineer, Gyornal
- [Niresh Shankar](https://himalayas.app/@nireshshankar): Founding Engineer, YeahScene
- [Md Murtuza ali](https://himalayas.app/@murtuza777): Lead Software Engineer (Co Founder), Mindalike Technologies
- [Sai Nitish](https://himalayas.app/@sainitish): AI Research & Interface Evaluation, Confidential AI Research Project
- [rohanraj shinde](https://himalayas.app/@rohanrajshinde): AI Researcher, TCS Research
- [Kumar R](https://himalayas.app/@kumarr): AWS Student builder Group Lead, Amazon Web Services (AWS)
