Skip to content
View sanjeev29's full-sized avatar
29 followers · 105 following

Achievements

Achievement: Pair Extraordinairex4Achievement: Pull Sharkx3Achievement: YOLOAchievement: Arctic Code Vault Contributor

Achievements

Achievement: Pair Extraordinairex4Achievement: Pull Sharkx3Achievement: YOLOAchievement: Arctic Code Vault Contributor

Block or report sanjeev29

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
sanjeev29/README.md

Hi, I'm Sanjeev 👋

Backend & distributed-systems engineer. I build the systems behind fast, reliable products — event-driven backends, real-time infrastructure, and the observability and CI/CD foundations teams depend on. Most of my career has been as an early or founding engineer, taking vague concepts to production without a playbook.

Recently I've been going deep on agentic AI and MCP tooling, bringing a reliability-first mindset to LLM systems: typed tool boundaries, verifier passes, and evals so agents actually hold up in production.

What I work on

  • Distributed & event-driven backends — Python, Go, Node.js, FastAPI, Django, GraphQL
  • Real-time systems — scaled livestream chat to 1,000+ concurrent viewers at sub-50ms delivery
  • Observability & reliability — SLO-driven on-call, Prometheus/Grafana RED metrics, cut high-severity MTTR by nearly half
  • Data & infra — PostgreSQL, Redis, Kafka, DynamoDB, AWS (Lambda, ECS, SQS), Docker, Terraform
  • Agentic AI — MCP servers, tool-use loops, RAG, typed Pydantic boundaries, agent evals

Recent open-source (mostly agent tooling)

  • scout-browser-agent — LLM agent that drives a real Chromium browser via a 4-layer perception cascade; typed action schema, 40 tests in CI.
  • AxiomLoop — research agent that maps who-claims-what across sources, with a verifier pass that catches fabricated citations. MCP + typed Pydantic.

Stack

Python · Go · TypeScript · FastAPI · Django · Node.js · GraphQL · PostgreSQL · Redis · Kafka · AWS · Docker · Terraform · Prometheus · Grafana · MCP

Reach me

Pinned Loading

  1. scout-browser-agent scout-browser-agent Public

    A minimal LLM agent that drives a real browser via a 4-layer perception cascade: HTTP text extraction, accessibility-tree legend, and set-of-marks vision.

    Python 1

  2. AxiomLoop AxiomLoop Public

    Agent that researches a question across arXiv & Google Scholar, records structured evidence, and writes a cited answer with a typed contradiction verifier.

    Python