Skip to content
View DevChiniwala's full-sized avatar
🎯
Focusing
🎯
Focusing
6 followers · 2 following

Achievements

Achievement: Pull Sharkx2Achievement: Pair Extraordinairex2Achievement: YOLOAchievement: Quickdraw

Achievements

Achievement: Pull Sharkx2Achievement: Pair Extraordinairex2Achievement: YOLOAchievement: Quickdraw

Highlights

  • Pro

Block or report DevChiniwala

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Content in all repositories owned by your account will be closed.
Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. Markdown is supported. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
DevChiniwala/README.md

01 / Flagship systems

Three public systems showing how I approach production AI: instrument the runtime, preserve changing state, and verify generated intelligence.

01 — ControlSurface / Agent production engineering

ControlSurface — agent observability, failure intelligence, regression evidence, and release gates

Production failures should become reusable engineering evidence. ControlSurface derives agent execution structure from traces, turns recurring failures into reviewed regression cases, and attaches deterministic evidence to release decisions.

Repository · Architecture · v0.1.0-rc.1
Python OpenTelemetry PostgreSQL ClickHouse Next.js Docker Compose

02 — MNEMOS / Temporal memory

MNEMOS — self-editing bi-temporal memory with six-signal retrieval

Memory should preserve both change and provenance. MNEMOS resolves new facts through explicit lifecycle operations, combines six retrieval signals, and keeps context bounded through consolidation, decay, and token-aware packing.

Repository · Package metadata · Documentation
Python MCP FastAPI Neo4j FAISS RAGAS

03 — CortexOS / Verified intelligence

CortexOS — multi-agent company research with deterministic source verification

Generated intelligence is useful when its claims can be inspected. CortexOS combines multi-agent research with a deterministic Rust verification path and a local Tauri application backed by SQLite.

Repository · Frontend manifest · Rust core
React TypeScript Tauri Rust Tokio SQLite

02 / Applied systems

04 — VYGOT / Adaptive learning

VYGOT — private team project exploring adaptive learning intelligence, with a public launch upcoming

VYGOT is a private team project exploring personalized learning intelligence. Product and implementation details remain limited until its public launch.

05 — HeyClara / Always-on personal agent

HeyClara — scheduled personal agent with persistent workspaces, memory, MCP tools, and multi-channel delivery

A personal agent should continue useful work between conversations. HeyClara runs scheduled jobs with persistent workspaces, consolidates memory in two stages, and exposes shared MCP tools across Telegram, Slack, voice, and SMS.

Repository · npm package · Documentation
Bun TypeScript Claude Agent SDK MCP PostgreSQL Zod

03 / System portfolio

Portfolio map of five independent systems across agent operations, temporal memory, source verification, adaptive learning, and personal automation

Each project is an independent system. The map describes the engineering surface each one owns without implying a shared runtime.

04 / Open-source engineering

I have contributed 70+ merged pull requests across AI infrastructure, developer tooling, retrieval systems, data platforms, and cross-platform reliability.

Selected contribution Engineering change
Reticle #1286 Preserved declared network URLs through browser action replay.
ChunkHound #419 Replaced embedding full-table scans with targeted SQL.
HelixDB #1164 Validated signed 64-bit values across generated SDKs.
DataMind #21 Redacted secrets from audit errors.

Browse all merged contributions →

05 / AI systems stack

Engineering stack across languages, models and AI, agents and context, runtime and product, data and memory, and infrastructure and reliability

06 / Engineering principles

Four engineering principles: reduce apparent complexity, design for evolution, shorten feedback loops, and turn failure into knowledge

The panel paraphrases ideas from A Philosophy of Software Design, Designing Data-Intensive Applications, Accelerate, and Site Reliability Engineering.

07 / Live GitHub telemetry

Daily generated public GitHub telemetry for DevChiniwala

Generated daily from GitHub's public REST API. Repository and contribution counts are scoped in the panel; language share uses owned, non-fork public repositories. Generator · Workflow

Dev Chiniwala contribution activity

08 / Connect

I build the systems around AI models: data, memory, evaluation, runtimes, and product delivery.

npx devcard0707

LinkedIn · X / Twitter · Email · GitHub

Pinned Loading

  1. ControlSurface ControlSurface Public

    Production engineering for AI agents — observability, reliability, incident intelligence, regression testing, and release gates.

    Python 1

  2. MNEMOS MNEMOS Public

    Self-editing, bi-temporal memory runtime for AI agents — 6-signal hybrid retrieval, Neo4j knowledge graph, FSRS cognitive decay, sleep consolidation. Not another vector DB wrapper.

    Python 1

  3. CortexOS CortexOS Public

    CortexOS is an AI-native intelligence operating system that uses autonomous agents to research companies, analyze market signals, verify insights, and generate actionable business intelligence in r…

    TypeScript 3

  4. HeyClara HeyClara Public

    HeyClara is an always-on AI assistant that runs in the background to automate your daily routines, scheduled tasks, and long-term memory. It communicates directly through messaging apps like Telegr…

    TypeScript 1