# Aman Hashmi

Aman Hashmi is a remote worker on Himalayas, Independent AI Output Evaluator at Specialized Practice, based in India.

Source: https://himalayas.app/@amanhashmi1

```text
LLM output evaluator practicing RLHF who compares model responses and audits for hallucinations, logic, and policy compliance.
```

| Detail | Value |
| --- | --- |
| Availability | Looking for a job |
| Location | India |
| Authorized to work in | India |
| Salary expectations | 70k+ INR |
| Job categories | AI Output Evaluator, AI Engineer |
| Skills | RLHF, LLM Evaluation, Preference Ranking, A B Testing, Rubric Based Scoring, Error Classification, LLM Output Assessment, Hallucination Detection, Instruction Adherence, Constraint Verification, Compliance Auditing, Source Text Fidelity, Semantic Consistency, Fallacy Detection, Logic |
| LinkedIn | https://linkedin.com/in/amaan-hashmi |

## What I'm looking for

```text
I'm seeking RLHF and model-evaluation roles.
```

## About

```text
In my independent practice as an AI Output Evaluator and Prompt Auditor, I compare multi-turn LLM responses side by side using a standardized rubric. I assess logical consistency, hallucinations, and policy compliance.

I analyze hallucinations to distinguish factual distortion and ungrounded claims from acceptable paraphrase, and apply formal logic to identify invalid deductions and fallacies. I write evidence-based justifications for model preferences, grounded in logical soundness, source fidelity, and instruction compliance.

Through independent study in formal logic, cognitive science, and critical theory, I’ve strengthened my approach to evaluating arguments and synthesizing conflicting sources without adding unsupported assumptions. My portfolio includes anonymized A/B evaluations covering source fidelity, fallacy identification, and policy compliance.
```

## Experience

### Independent AI Output Evaluator

Specialized Practice

```text
Evaluated dozens of multi-turn LLM responses side by side (A/B) using a standardized rubric for logical consistency, hallucination detection, and policy compliance. Tracked four error categories and performed root-cause analyses on hallucinations.
```

Skills: RLHF, Side-By-Side-Evaluations, Rubric-Based-Scoring, Hallucination-Detection, Fallacy-Detection, Policy-Compliance, Logic, Error-Classification

### Analytical Research & Critical Reasoning Studies

Independent Study

```text
Engaged in structured, self-directed study across formal logic, cognitive science, and critical theory, sharpening premise evaluation, counter-argumentation, and conceptual stress-testing. Synthesized conflicting texts from multiple domain sources.
```

Skills: Logic, Critical-Thinking, Argument-Analysis, Research, Synthesis

## Education

### BIRT

Bachelor of Technology, Computer Science and Engineering. 2023 - Present

```text
Pursuing a Bachelor of Technology in Computer Science and Engineering, expected to graduate in 2026.
```

## Tech stack

- Python
- Docker

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

## People also viewed

- [Udaykiran Suryadevara](https://himalayas.app/@udaykiransuryadevara): AI Operations Associate, Deccan AI
- [Muskan Singh](https://himalayas.app/@muskansingh2): AI Evaluation Specialist, Independent
- [Sai tarun Kadari](https://himalayas.app/@saitarunkadari): Senior AI Quality Analyst, Independent / Freelance
- [Tejas Mohare](https://himalayas.app/@tejasmohare): LLM Evaluation Specialist, Outlier
- [Kushal Jaiswal](https://himalayas.app/@kushaljaiswal): AI Trainer / LLM Evaluator, Outlier AI (Scale AI)
- [Xrs Here](https://himalayas.app/@sumanmahato): AI Output Evaluation & Data Annotation, Heritage Institute of Technology
- [Srin Srinivas](https://himalayas.app/@srinsrinivas): AI/LLM Analyst, Innodata
- [Ankur Roy](https://himalayas.app/@ankurroy): AI Trainer & Evaluation Specialist, Outlier.ai
- [Nagendra Madineni](https://himalayas.app/@nagendramadineni): AI Agent Evaluation & Prompt Engineering, Turing
- [Aman Yadav](https://himalayas.app/@amanyadav17): AI Response Quality Evaluator, Outlier.ai
