Skip to content
View Kcodess2807's full-sized avatar
💭
vibing :)
💭
vibing :)

Block or report Kcodess2807

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. 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
Kcodess2807/README.md

Terminal: Arush Karnatak, AI engineer. Focus: agent infrastructure and memory for agents. Shipped: Graphsight and Saadho. Learning: inference engineering.

Terminal tree of the agent stack: orchestration shipped, memory main focus, payments some, inference learning.

Featured work

Graphsight: project memory for coding agents

Terminal: remembering a fact, replacing it, and the facts the model then sees. Eval: no memory 5 of 22, episodic 14 of 22, graph memory 16 of 22.

It turns a repo's GitHub history into a typed graph, then lets an agent add the team facts GitHub doesn't hold. A tracer shows what the agent retrieved and what it actually used.

  • Memory an agent can correct. A changed fact ends the old one and keeps it as history. A retracted fact stops being recalled.
  • A source on every fact. The model is told which facts came from GitHub and which an agent was told.
  • Measured, with the misses published. Tasks are written before any run and graded by exact match. The eval write-up lists what failed as well as what passed.

Python LadybugDB LangGraph MCP FastAPI React  Code · PyPI · Site · Eval

Saadho: AI/ML interview prep, day by day

Terminal sketch of Saadho: a role, an interview date and daily hours produce a day-by-day plan, plus a voice practice room with four AI colleagues.

Give it a role, an interview date and your daily hours, and it builds the plan. Live in production.

  • A plan, in order. 300+ vetted resources sequenced by prerequisite, with what to skip.
  • Company guides. How 200+ companies interview, round by round, with sources.
  • A voice practice room. A group discussion with four AI colleagues in English or Hinglish, then a scorecard that quotes your own words.

Next.js Supabase LLM agents Realtime voice  Live site

Reading for open-source contributions: inference

I'm working down the stack, from agents to how a model is actually served. These are the codebases I'm reading with the aim of contributing.

Repository What I'm reading it for
vllm-project/vllm Continuous batching, PagedAttention, and the serving scheduler
ggml-org/llama.cpp Quantization formats (GGUF) and CPU/GPU inference in plain C++
sgl-project/sglang Prefix caching with RadixAttention and structured-output serving
NVIDIA/TensorRT-LLM Engine building and GPU-side optimization
ModelTC/LightLLM A lightweight Python serving framework: async processes and dynamic batching
skyzh/tiny-llm Building an inference engine from scratch, to learn the internals
modular/llm-inference-handbook The reference I read alongside the code

Notes and experiments: inference-engineering0to100.

More work

Project What it is
SmartFlow A reinforcement-learning traffic-signal controller. Cuts average waiting time about 55% against a fixed-time controller, and extends to a multi-agent 2x2 grid with emergency-vehicle preemption. SUMO PPO/DQN · Code
Bodhi A voice-first AI mock interviewer: an adaptive pipeline, RAG over 10K+ chunks, and real-time proctoring. LangGraph pgvector YOLOv8
PADHO Agents with hierarchical orchestration and RAG, in pure Python with no framework.
venvctl A single-binary Windows CLI that manages Python virtual environments across PowerShell, CMD and Git Bash. Go
ReAct-CLI A small command-line implementation of the ReAct loop: reason, act, observe.

Background

  • Built ML pipelines processing 10,000+ aerial images a day at Kanchan Drones.
  • Built a real-time exam-proctoring system holding sub-200ms latency on GCP.
  • Shipped a policy-aware RAG chatbot that answers in under a second.
  • Taught an 8-week NLP course to 20+ students.

Find me

LinkedIn X Portfolio Email

Pinned Loading

  1. beetlejusse/SMARTFLOW beetlejusse/SMARTFLOW Public

    SMARTFLOW leverages advanced vision-based technologies, including face emotion analytics, recognition, and automated blurring systems, while utilizing Indian driving datasets to detect vehicles, po…

    Python 5 4

  2. Bodhi Bodhi Public

    Forked from SOGeKING-NUL/Bodhi

    Building a voice-first, low-latency AI Mock Interviewer

    Python

  3. graphsight graphsight Public

    Project memory for coding agents: GitHub history as a typed graph, plus a tracer that shows what your agent retrieved and what it used.

    Python 5 1

  4. venvctl venvctl Public

    Single-binary Windows CLI that manages Python virtual environments: one on/off in PowerShell, CMD and Git Bash; find, unlock, repair and prune venvs. uv creates, venvctl manages.

    Go 1 1