“All the code that’s fit to ship.”

Resume PDF (opens in a new tab)

Aansh Singh, The Aansh Times

Los AngelesSoftware EngineerUCLA Computer ScienceVol. 1, No. 1

Aansh Singh

Software Engineer / UCLA Computer Science

Aansh Singh at Rockefeller Center, New York.

Pragmatic, system-minded.

Hi, I’m Aansh Singh. A computer science student at UCLA building developer tools, AI systems, and reliable backend software.

  • Software Engineer
  • Developer Tools
  • AI Systems
  • Reliable Backend Software
  • UCLA · Los Angeles, CA.

Projects

Developer Tools

Repository Tool Cuts LLM Context 97.5%

CodeGraph indexes a repository’s files, functions and calls so an AI agent retrieves only the context a question needs.

Explore codebases through interactive graphs and question-specific retrieval.

Structure & source. Inspect source on demand and trace direct callers and architectural communities.

I designed and built CodeGraph independently during summer 2026, before the school year began. My work spans the full system: repository ingestion and parsing, the graph and retrieval pipeline, the API, and the interactive dashboard.

97.5% less LLM context

≈737K → ≈18K tokens/query; supported source vs. retrieved tool context

Built with

  • React / TypeScript
  • Python / FastAPI
  • Tree-sitter
  • Neo4j
  • LangGraph
Continued: CodeGraph case study Source: CodeGraph on GitHub

Ingest: the tiangolo/fastapi repository, in full from its default branch or as the files a push or pull request changed at that commit, merges into one stream of supported source files: the fastapi package, its tests and documentation examples, and the documentation’s JavaScript. Files such as README.md and uv.lock are skipped, and every source file is token-counted.

Parse: Tree-sitter reads Python, JavaScript, TypeScript, Go and Java. In fastapi/dependencies/utils.py it extracts the function analyze_param, its call to the external is_typealiastype, and its call to create_model_field.

Graph: inside the tiangolo/fastapi boundary, files define functions and functions call functions; calls to code outside the repository, such as run_in_threadpool and deepcopy, are marked external. Every function carries an embedding, and clustering groups functions into labeled subsystems, shown here as Request Routing, Dependency Injection and OpenAPI Schema.

Answer: asked “What breaks if I change create_model_field?”, a Claude agent finds create_model_field by semantic search, walks to its 3 callers (analyze_param and _get_body_field, and _populate_api_route_state in the routing subsystem), and reads the Dependency Injection subsystem, repeating until it has enough context.

Context sent to the model: ≈737K tokens for the full repository against ≈18K tokens of retrieved context, 97.5% fewer.

Conceptual view. Calls excerpted from tiangolo/fastapi; subsystem names illustrative; bars to scale.Graphic

Distributed Systems

Go Workers Scale 15.84× Under Load

TaskForge coordinates durable background tasks through PostgreSQL, with concurrent workers, retries and crash recovery.

Coordinate durable background work with concurrent Go workers, atomic claims, retries, and crash recovery.

Recovery with history. Keep numbered attempts for completion, retry, and crash recovery.

I designed and built TaskForge independently during summer 2026, before the school year began. My work spans the full system: the API and console, PostgreSQL task and attempt model, Go workers and schedulers, observability, and benchmark harness.

15.84× worker scaling

1 → 16 workers on synthetic 50 ms waits; local Docker benchmark

Built with

  • Go
  • PostgreSQL
  • Docker Compose
  • Prometheus
  • Grafana
Continued: TaskForge case study Source: TaskForge on GitHub
  1. Queue

    Durable tasks

  2. Workers

    Atomic claims

    • Completion
    • Retry / backoff

Due retries return to the queue

Conceptual view. Task coordination and retry flow.Graphic

Business & Campus

Experience

  1. Software Engineer, Bruin Underwater Robotics (BUR) @ UCLA

    • Developed Python/YOLOv11 perception models for autonomous RoboSub and automated a Unity pipeline generating 3,000+ labeled training images.
  2. Operations Team Member, exploretech.la

    • Coordinated event logistics and procurement for an annual STEM outreach event serving ~500 high school students across Greater Los Angeles.
    • Partnered with ASUCLA Catering and external vendors to source catering, event supplies, and custom merchandise, ensuring materials were ready for event.
  3. Software Engineering Team Lead Intern, D-Tech

    • Led 5 interns designing Flippper, an ePaper education platform, translating requirements into user flows and functional specifications.
    • Directed UI/UX prototypes and visualizations to validate workflows and explain system behavior.
  4. Advanced Academic Ambassador - Projects Chair, Round Rock Independent School District (RRISD)

    • Led and mentored 30 ambassadors developing projects and educational presentations for the student body.
    • Represented over 34,000 students, delivering presentations and answering questions at conferences and college fairs.
  5. Instructor, CompuChild

    • Taught Python and Scratch to middle school students through hands-on coding projects.
    • Guided line-following robot and solar-car projects, helping students debug code and devices.

Notice

Education

University of California, Los Angeles (UCLA)

Henry Samueli School of Engineering

Bachelor of Science in Computer Science

Classifieds

Contact

Wanted

Let’s build something useful.

Have a project or an idea to discuss? Get in touch.

Seeking
Summer 2027 software engineering internships, plus off-cycle (fall or winter) co-ops and part-time roles.
Work authorization
Authorized to work in the United States; will not require visa sponsorship.
Location
Based in Los Angeles, CA (UCLA) and Austin, TX; open to relocating anywhere in the U.S. and to remote work.

Reply to singhsitanshu@ucla.edu