# Zig Projects at AI Tinkerers

> Canonical HTML: https://aitinkerers.org/technologies/zig
> Markdown URL: https://aitinkerers.org/technologies/zig.md
> Technology record last updated: 2026-02-28T01:02:08Z
> Generated: 2026-09-21T01:40:50Z

Zig is a modern systems programming language and toolchain: it delivers C-level performance, zero-dependency cross-compilation, and powerful compile-time metaprogramming via its `comptime` feature.

Zig is a general-purpose, statically typed systems language designed by Andrew Kelley to improve upon C: it prioritizes robustness, optimality, and reusability. The toolchain acts as a zero-dependency C/C++ compiler, providing seamless cross-compilation for dozens of targets right out of the box. Its core differentiator is `comptime` (compile-time code execution), which handles generics, reflection, and metaprogramming without a preprocessor. For memory, Zig uses manual management with explicit allocators, eliminating hidden control flow or memory allocations. Companies like Uber and TigerBeetle leverage Zig for high-performance, low-level software.

- Official technology site: https://ziglang.org
- Public AI Tinkerers demos and talks: 3
- Result page: 1 of 1

## Recent Public Talks and Demos

### [Evolutionary Harness: A Coding Agent That Breeds, Judges, and Promotes Its Own Promptsz](https://singapore.aitinkerers.org/talks/rsvp_sEM3ge8rdCE)

Codegraff is an open-source coding-agent harness — a single ~2.7 MB dependency-free Zig binary — that treats system prompts as genomes and evolves them: it mutates its own persona, runs the variants in tournament, has an independent LLM judge score them against a deterministic eval, and promotes the winner to become the root prompt of future sessions, all on a local first, HMAC-signed, SHA-addressed learning store. Live demo (code + working system, no slides): - A running agent session next to `tail -f .graff/traces/.jsonl` — the harness's own brain: per-turn API latency, cache reads, and tool timings that the agent itself is prompted to read for self-debugging. - A fleet tournament: one workflow phase fans out 3 prompt variants of the same task in parallel, an independent read-only judge subagent scores each 0–100, and signed fitness telemetry feeds a MAP-Elites archive keyed by (niche × provider × eval-set hash). - The self-improvement loop closing itself: at session end the workspace auto-bootstraps `.graff/learn/` (immutable content-addressed genomes/evidence/runs, only mutable commit point an atomically-replaced `refs/active.json`) and queues a background trial — mutate → paired eval vs. parent → statistical gates → hidden holdout → promote or reject. - The repo itself: `src/fleet.zig`, `src/learn_tournament.zig`, `src/subagent.zig` (judge), `src/agent_eval.zig` (eval gate).

- Event context: AI Tinkerers x Redis: Built for Speed — 2026-08-05 — Singapore
- Public talk page: https://singapore.aitinkerers.org/talks/rsvp_sEM3ge8rdCE

### [CodeDB: Building a Code Intelligence Server That Cuts Agent Token Usage](https://singapore.aitinkerers.org/talks/rsvp_YyFDIYV6hPs)

CodeDB is a code intelligence server for AI agents that indexes a codebase once and serves structured answers through MCP, HTTP, and CLI instead of forcing models to repeatedly scan raw files. In this demo, I show how CodeDB uses structural indexing, trigram search, inverted word lookups, dependency graphs, and snapshots to help agents find only the most relevant code context with much lower latency and dramatically fewer tokens.

- Event context: AI Tinkerers Singapore: The Agentic Future & Dev/Eng Workflows — 2026-04-21 — Singapore
- Public talk page: https://singapore.aitinkerers.org/talks/rsvp_YyFDIYV6hPs

### [Zig in AI?](https://singapore.aitinkerers.org/talks/rsvp_rD8SYTfKDUo)

It will be a talk about why Zig might make for an interesting language to use for production deployment and development of AI application instead of Python.

- Event context: AI Tinkerers Singapore: Meetup - September 16th, 2024 — 2024-09-16 — Singapore
- Public talk page: https://singapore.aitinkerers.org/talks/rsvp_rD8SYTfKDUo

## Related Technologies

- [Anthropic API](https://aitinkerers.org/technologies/anthropic-api) ([Markdown](https://aitinkerers.org/technologies/anthropic-api.md)) — 66 public demos
- [dependency graphs](https://aitinkerers.org/technologies/dependency-graphs) ([Markdown](https://aitinkerers.org/technologies/dependency-graphs.md)) — 1 public demo
- [GitHub](https://aitinkerers.org/technologies/github) ([Markdown](https://aitinkerers.org/technologies/github.md)) — 74 public demos
- [HTTP](https://aitinkerers.org/technologies/http) ([Markdown](https://aitinkerers.org/technologies/http.md)) — 14 public demos
- [inverted word indexing](https://aitinkerers.org/technologies/inverted-word-indexing) ([Markdown](https://aitinkerers.org/technologies/inverted-word-indexing.md)) — 1 public demo
- [JSON-RPC](https://aitinkerers.org/technologies/json-rpc) ([Markdown](https://aitinkerers.org/technologies/json-rpc.md)) — 2 public demos
- [MAP-Elites](https://aitinkerers.org/technologies/map-elites) ([Markdown](https://aitinkerers.org/technologies/map-elites.md)) — 2 public demos
- [MCP](https://aitinkerers.org/technologies/mcp) ([Markdown](https://aitinkerers.org/technologies/mcp.md)) — 129 public demos
- [OpenAI API](https://aitinkerers.org/technologies/openai-api) ([Markdown](https://aitinkerers.org/technologies/openai-api.md)) — 520 public demos
- [OpenTelemetry](https://aitinkerers.org/technologies/opentelemetry) ([Markdown](https://aitinkerers.org/technologies/opentelemetry.md)) — 11 public demos
- [Python](https://aitinkerers.org/technologies/python) ([Markdown](https://aitinkerers.org/technologies/python.md)) — 662 public demos
- [structural parsing](https://aitinkerers.org/technologies/structural-parsing) ([Markdown](https://aitinkerers.org/technologies/structural-parsing.md)) — 1 public demo
- [trigram indexing](https://aitinkerers.org/technologies/trigram-indexing) ([Markdown](https://aitinkerers.org/technologies/trigram-indexing.md)) — 1 public demo