H.HARSH SINGH

ARTIFICIAL INTELLIGENCE · SOFTWARE · RESEARCH INTERESTS

Curious about
how machines
think.

I am interested in artificial intelligence, particularly how intelligent systems reason, plan, explore alternatives, and verify their outputs. I explore these questions through software projects and technical experimentation.

RESEARCH-MINDEDENGINEERING-LEDBUILDING IN PUBLIC
↳

Long-term focus: building a clearer understanding of reasoning, computation, and intelligent systems.

AI REASONING✳MACHINE LEARNING✳AGENTIC SYSTEMS✳QUANTITATIVE SYSTEMS✳HPC & COMPUTATION✳ AI REASONING✳MACHINE LEARNING✳AGENTIC SYSTEMS✳QUANTITATIVE SYSTEMS✳HPC & COMPUTATION

I learn by building.
I build to understand.

I am interested in artificial intelligence and the foundations of intelligent problem-solving. My primary research direction is AI reasoning: how systems represent complex tasks, explore alternatives, use additional computation when it is useful, and verify their results.

I explore these interests through software projects involving AI systems, technical workflows, and interactive tools. I aim to describe each project accurately and distinguish implemented work from ideas or planned features.

[ CURRENT APPROACH ]Understand the problem. Build the smallest useful system. Verify the result. Document the limits.

Clarity before
complexity.

01 / UNDERSTAND

Start with the problem

Clarify the question, constraints, users, and evidence before choosing an implementation.

02 / BUILD

Make ideas inspectable

Turn concepts into focused tools, reproducible workflows, and systems that can be examined.

03 / VERIFY

Be precise about results

Distinguish tested behavior from assumptions, prototypes, open questions, and unverified work.

Technical skills
and tools.

A focused selection of languages, AI frameworks, agentic-development tools, and engineering workflow tools. See the complete toolkit for project-linked skill libraries and visual references.

Explore the technical toolkit ↗
LANGUAGES

Python · JavaScript · Java · SQL

AI, MACHINE LEARNING & RESEARCH

PyTorch · TensorFlow · NumPy · Pandas · MetaPipeline · Scikit-learn

AGENTIC DEVELOPMENT

OpenAI Codex · Claude Code · Gemini CLI · Agent Skills · MCP · Multi-agent workflows · Prompt & context evaluation

ENGINEERING & WORKFLOW

Git · GitHub · GitHub Actions · Linux

From one-pass answers
to deliberate computation.

I'm interested in how models can structure difficult tasks, decide when more computation is useful, explore alternatives, and verify their outputs. These are directions I am studying and exploring—not claims of completed scientific findings.

Read the research direction ↗

05 / CONTACT & COLLABORATION

Let's work on
good questions.

I'm interested in thoughtful discussions around AI reasoning, machine learning, model evaluation, agentic engineering, and open-source projects.