Srijan Shukla
Back home

Srijan Shukla

Bengaluru, India | github.com/srijanshukla18 | linkedin.com/in/srijanshukla18 | srijanshukla18@gmail.com | srijanshukla.com | +91-9686039480

Short resume

Senior SRE turned AI builder. I build reliable agent systems, developer tooling, and production infrastructure.

Over 8 years in infrastructure and reliability across AWS, Azure, and GCP, with hands-on ownership of large-scale systems, incident response, observability, and cost optimization. I have helped cut AWS spend by about $800K per year, reduced storage costs by about $8900 per month, operated ingest systems peaking at 350M events per minute, and scaled data pipelines beyond 10 million packets per day.

In parallel, I build applied AI products with real users. My open-source MCPs and AI developer tools have earned 140+ GitHub stars. Projects include xray for progressive code intelligence, claude-memory-viz for memory visualization, contextgraph for decision auditing in agents, kubectl-smart for Kubernetes signal prioritization, and multiple RAG and agent systems across legal search, offline knowledge retrieval, IAM, and voice interfaces.

I bring founder energy with operator discipline: 3x founder, strong in Python and Go, comfortable from zero-to-one product work through production hardening, platform reliability, and developer experience.

Long resume

Builder who ships. Senior SRE with 8+ years building and operating multi-cloud systems across AWS, Azure, and GCP, and 9 years building products across AI tooling, infrastructure, and startups.

Created open-source MCPs and AI developer tools with 140+ GitHub stars. Serial entrepreneur with a 3x founder background. Ship 0 to 1 products fast with SRE-grade reliability.

Strong in Python, Go, RAG, embeddings, Kubernetes, Terraform, observability, incident response, and build automation.

What I do

AI agent systems, infrastructure engineering, reliability, and shipping practical developer tooling.

Focus areas

  • LLM workflows and tooling
  • Distributed systems & platform reliability
  • Developer ergonomics and observability
  • Product-minded infrastructure decisions

Experience

Independent AI Builder | 2023 - Present

Build and ship applied-LLM products and infra with real adoption. Combine RAG, KG search, and reliable deployment.

  • claude-memory-viz: Memory visualization for Anthropic's Claude MCP; embedding + clustering visualizer; 95 GitHub stars.
  • xray: MCP providing progressive code intelligence for AI assistants via ast-grep; 43 GitHub stars.
  • contextgraph: Decision audit ledger for AI agents; captures why as queryable data; full-stack with SDK, server, UI.
  • alpha: IAM policy rightsizing agent with AI-powered risk signals and instant rollback.
  • logsieve: Log deduplication sidecar using Drain3 algorithm; production-ready with Helm charts; Go.
  • wiki-in-a-box: Offline Wikipedia with hybrid no-index RAG; local-first knowledge retrieval.
  • ita-kg: Income Tax Act Knowledge Graph + RAG system for legal lookup.
  • murmur: Voice interface for Claude Code/Codex CLI using whisper.cpp with Metal acceleration.
  • global-publish: Intelligent content generator adapting writing for 12 social platforms.
  • kubectl-smart: CLI transforming Kubernetes debugging to intelligent signal prioritization.

SRE-III | SteelEye | Jul 2023 - Present

  • Primary SRE for a tier-1 client: deployments, prod debugging, and maintenance.
  • Led infra for a major POC and implemented Azure RBAC with managed identities, including SDK-level permission deep dives.
  • Helped cut AWS spend by about $800K per year as a core infra contributor on a cloud-agnostic migration.
  • Partnered on Elasticsearch snapshot archival; infra setup now saves about $8900 per month in storage.
  • Implemented multi-tenant monitoring with the Grafana stack: Prometheus, Loki, Tempo, and Grafana, including dashboard approach and data-retention strategies.
  • Co-built an ops automation platform using Go and Temporal for long-running maintenance workflows.
  • Standardized environment bootstrapping by packaging Ansible as a Kubernetes Job via Helm, covering secrets, RBAC, and base config.
  • Mentored juniors and supported incident management within the team.

Software Engineer - Sr. SRE | Last9 | Dec 2021 - Apr 2023

  • Owned infra for a metrics and event ingest system peaking at 350M events per minute; tuned throughput, reliability, and cost.
  • Led performance tuning and capacity planning on AWS and GCP; achieved about 20% cost reduction via rightsizing and workload placement.
  • Guided Kubernetes orchestration choices, removed bottlenecks, and documented failure modes and FMEA for critical paths.
  • Streamlined customer onboarding via VPC Peering and AWS PrivateLink; shipped metrics and log pipelines end-to-end.
  • Hardened backups and security controls, mapped to SOC 2, and improved incident readiness and recovery drills.

Sr. Software Engineer (Infrastructure) | TNG Innovation Labs | Jan 2019 - Nov 2021

  • Scaled data pipelines to 10 million+ packets per day; owned infra, observability, and release pipelines.
  • Drove system design and also acted as interim tech lead and database engineer.
  • Led transition from HTTP to MQTT, significantly cutting costs and improving performance.
  • Migrated MySQL from self-hosted to AWS RDS with near-zero downtime and instituted backups and disaster recovery.

Founding Engineer | Peekstreets | Feb 2024 - Jul 2025

Built AI infrastructure and backend for a public-equity research SaaS.

  • Designed RAG over multi-source datasets with retrieval caching and query planning; combined embeddings, metadata filters, and ranking.

Founding Engineer | LiQR (QR-Menu Startup) | Jul 2020 - Sep 2020

Delivered MVP and initial launch during COVID.

  • Built a Flutter app with a Python backend on AWS using Terraform; shipped to multiple restaurants in month one.

Founder & CTO | Deventree Solutions | May 2016 - Dec 2018

Built telematics platform and cost-efficient location services.

  • Scaled to about 20k devices and about 4M messages per day; a self-hosted reverse geocoder cut infra cost by about 80%.
  • Owned frontend, backend, infra, and database.
  • Shipped multiple products using Android, Rails, Node.js, Angular, and React Native.
  • Managed development, infra, and architecture.
  • Led a small engineering team.
  • Solved a complex timetabling problem using a genetic algorithm.

Projects and side work

  • claude-memory-viz: Claude Memory MCP Visualizer.
  • xray: Progressive code intelligence and navigation for AI assistants through structural code analysis using ast-grep.
  • kubectl-smart: A CLI tool that transforms Kubernetes debugging from reactive noise filtering to intelligent signal prioritization.
  • global-publish: Intelligent content generator adapting writing for 12 social platforms.
  • murmur: Voice interface for Claude Code and Codex CLI using whisper.cpp with Metal acceleration.
  • contextgraph: Decision audit ledger for AI agents with SDK, server, and UI.
  • alpha: IAM policy rightsizing agent with AI-powered risk signals and instant rollback.
  • logsieve: Log deduplication sidecar using the Drain3 algorithm with Helm charts.
  • wiki-in-a-box: Offline Wikipedia with hybrid no-index RAG.
  • ita-kg: Income Tax Act Knowledge Graph + RAG system for legal lookup.

Core skills

  • Applied AI: RAG, embeddings, LLM evaluation, prompt and agent design, MCP development, vector DBs including FAISS and pgvector, Neo4j and Cypher, ast-grep, whisper.cpp.
  • Infrastructure: Kubernetes, Docker, Terraform, GitHub Actions, GitOps with Flux and Helm, AWS including EC2, EKS, Lambda, S3, IAM, VPC, Azure, GCP, eBPF.
  • Programming languages: Python, Go, C, Bash, JavaScript, SQL.
  • Data and ops: PostgreSQL, Redis, Elasticsearch, Kafka, Prometheus, Grafana, Loki, Tempo, cost optimization, incident response.
  • Cloud providers: Proficient in AWS including EKS, EC2, RDS, S3, EFS, SNS; comfortable with Azure and GCP.
  • Infrastructure and tools: Linux, Kubernetes, Terraform, Ansible, Prometheus, VictoriaMetrics, Grafana, Docker, NGINX, Helm, Flux, HAProxy, RabbitMQ.
  • Databases and caches: MySQL, MongoDB, PostgreSQL, Redis, Memcached.

Education and certifications

  • B.Tech in Computer Science and Engineering, Christ University, Bangalore | 2013-2017
  • B.Tech in Computer Science, Christ University, Bangalore | GPA: 3.27/4 | 2013-2017
  • Certified Kubernetes Administrator (CKA), Linux Foundation | Sep 2021
  • Certified Kubernetes Administrator (CKA), Linux Foundation | Credential ID: LF-m5ro376g4g | Sep 2021
  • Y Combinator Startup School | Jun 2017
  • Startup School Online, Y Combinator | Credential ID: 10740183 | Jun 2017

Talks

  • Lightning Talk: Slow down Disk I/O. Flash talk on how to stop rm from nuking your SSD, a deep dive into I/O throttling with rsync, ionice, and why cgroups v2 finally gets it right.