Skip to content
View harjothkhara's full-sized avatar
🎯
Focusing
🎯
Focusing

Block or report harjothkhara

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.

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
harjothkhara/README.md

Hi, I'm Harjoth 👋

X @harjothk LinkedIn

☕ 👨🏽‍💻 🇬🇧 I like building software with amazing people and contributing to open source projects. ex SWE @clipboardhealth · MS&E @stanford @ecorner

Contributing to

  • OpenClaw GitHub stars · 32 shipped - migrate MCP flags in included config files, preserve custom Ollama memory provider endpoints, accept local file URLs for managed images, safely format circular provider stream errors, keep Slack health probes within their timeout, preserve finalized Mattermost replies after tool warnings, correct invalid Telegram 401 recovery guidance, keep browser tabs between persistent cron runs · all merged PRs → · also 5 issues reported, incl. a P1 data-loss bug (#103363)
  • Python / CPython GitHub stars · 11 shipped - respect VIRTUAL_ENV_DISABLE_PROMPT in activate.bat (+ 3.13/3.14 backports), add a translate_line_endings parameter to imaplib APPEND (stop rewriting CR/LF), clarify venv target-directory reuse, document PyBytesWriter_GetData pointer validity, update the Windows NuGet download URL, fix a data race in readline.get_completer() and get_pre_input_hook()
  • Go GitHub stars · 10 shipped - modernize analyzer landed in x/tools, validate trailers when writing HTTP requests and responses, document PCG and ChaCha8 tradeoffs in math/rand/v2, update RFC 9106 argon2 parameter recommendations in x/crypto, record a test's failure for -failfast after cleanup runs (testing), reject invalid comment tokens in encoding/xml, skip the cgo compiler version probe under -toolexec (cmd/go), emit coverage for selected build dependencies (cmd/go)
  • NemoClaw GitHub stars · 10 shipped - make e2e snapshot commands inference-hermetic, never capture or restore OpenClaw runtime auth state in snapshots, centralize e2e workflow plan output, document the hermes API bearer token lifecycle and retrieval, scope the agent version probe to the sandbox's gateway, treat preset picker stdin EOF as cancellation, fail config set when the new-key prompt reaches EOF, correct the NEMOCLAW_CONFIG_ACCEPT_NEW_PATH docs and document config set
  • Hermes-Agent GitHub stars · 7 shipped - skip symlinked stage2 chown targets in the Docker build, recycle idle and long-lived MCP stdio servers with lazy reconnect, route typed configured models off the retired openai-codex provider, gitignore the Desktop bootstrap marker so updates stop autostashing it, preserve Desktop sidebar workspace targets across new drafts, avoid a one-shot SIGABRT during CLI teardown, decay protected summaries after restart
  • CodexBar GitHub stars · 5 shipped - isolate Codex OAuth quota requests per account, cool down Claude keychain pre-alerts, correct usage accounting for restarted Codex subagents, correct Ollama session token accounts, fix Kimi quota pace metadata
  • Astryx GitHub stars · 2 shipped - add a hermes agent-docs target to astryx init, parse CRLF blog frontmatter in the docsite
  • vLLM GitHub stars · 2 shipped - don't double-count local prefix-cache stats on scheduling retries, document NVFP4 GEMM kernel selection and the Marlin weight-only fallback
  • Node.js GitHub stars · 1 shipped - load inspector network-tracking modules eagerly, enforce readonly state on open vfs handles, fix TLS session reuse with provided sockets, clarify PEM format for signing keys in docs
  • Kubernetes GitHub stars · 1 shipped - clarify what setting .spec.os.name does in the Pod OS docs
  • Qwen Code GitHub stars · 1 shipped - fall back to the system ripgrep when the bundled binary cannot run

dino

Pinned Loading

  1. openclaw/openclaw openclaw/openclaw Public

    Your own personal AI assistant. Any OS. Any Platform. The lobster way. 🦞

    TypeScript 385k 80.8k

  2. NousResearch/hermes-agent NousResearch/hermes-agent Public

    The agent that grows with you

    Python 222k 42.7k

  3. golang/go golang/go Public

    The Go programming language

    Go 135k 19.2k

  4. vllm-project/vllm vllm-project/vllm Public

    A high-throughput and memory-efficient inference and serving engine for LLMs

    Python 87.6k 20k

  5. python/cpython python/cpython Public

    The Python programming language

    Python 74k 35.1k

  6. NVIDIA/NemoClaw NVIDIA/NemoClaw Public

    Run agents like Hermes, LangChain Deep Agents, and OpenClaw more securely inside NVIDIA OpenShell with managed inference

    TypeScript 22k 3k