What stack does Sairam work with?

Python, Go, TypeScript, AWS (Bedrock, Polly, Transcribe), LLM orchestration (crewAI, multi-agent), RAG, ReAct pipelines, Redis, SSE, and React/Next.js on the frontend.

Is Sairam open to new roles?

Yes — actively looking for Backend, GenAI, and Full-Stack engineering roles. Remote or Hyderabad, India.

What has Sairam built in production?

Pensieve (multi-agent LLM orchestration engine, 2K+ daily users), AAVA Code (AI coding plugin for VS Code, 3K+ daily users), Execution Engine (prompt-driven artifact generation, 65%→85% first-pass acceptance), Wireframe Generator (GenAI UI prototyping, 500+ daily users), and Prompt-to-React (code generation, 2K+ developers).

Where can I schedule a call with Sairam?

Book a 30-minute intro call at https://calendly.com/sairamugge/30min

What open-source projects has Sairam built?

MindForge (agentic Claude Code framework, 1193+ commits), Agent-Forge (self-improving agent infrastructure), ContextOS (AI agent intelligence layer), Graph-Forge (code intelligence platform), and others. See https://anvilry.vercel.app/projects.

> GenAI & Backend Engineer @ Ascendion

I build production multi-agent LLM systems and the event-driven backends behind them.

I co-built Pensieve and architected the AAVA Code backend at Ascendion, and build open-source AI infrastructure in the open.

See my work Résumé

2K+: daily usersPensieve
3K+: daily usersAAVA Code · 5+ clients
11: open-source reposAI infrastructure

// production work @ Ascendion

Flagship systems I helped build

Co-built · production-hardened

Pensieve

AI Process-Orchestration Engine · Ascendion

A multi-agent LLM orchestration engine with human-in-the-loop approval gates, real-time streaming, and governed LLM routing across cloud providers.

2K+: daily users across domains
HITL: approval gates
Multi-cloud: governed LLM routing

Co-built · architected the backend

AAVA Code

AI Coding Plugin for VS Code · Ascendion

Re-architected a single-agent prototype into a multi-agent orchestration system (Main-Agent + sub-agent via crewAI flows) powering an AI coding plugin for VS Code.

3K+: daily users
5+: client environments
150+: skills · 40+ tools · ~60 commands

Co-built · production-ready

Wireframe Generator

GenAI Wireframe Generator · Ascendion

A GenAI-powered wireframe generator that converts product inputs (PRDs, sketches, prompts) into production-ready UI artifacts, built as a full vertical slice within Experience Studio over an SSE real-time architecture.

40%: fewer UX iteration cycles (5 rounds → 3)
60%: faster prototyping
500+: users daily

Co-built · production-ready

Prompt-to-React

Prompt-to-React Code Generation · Ascendion

A prompt-to-React code generation system that turns generated wireframes into production-ready modular components and routing logic.

50%: less frontend dev time (2 days → 1 per feature)
55%: less manual coding effort
2K+: developers
50+: teams

Co-built

Execution Engine

Prompt-Driven Execution Engine · Ascendion

A prompt-driven execution engine using LLM-based agent orchestration with RAG and ReAct pipelines that automates generation of structured product artifacts.

1.5h → 15m: workflow planning time
65% → 85%: first-pass acceptance
1.5K+: users

// open-source AI infrastructure

Things I build in the open

MindForge

Agentic-intelligence framework for Claude Code.

Command-driven multi-agent orchestration for Claude Code — 174 commands, 154 subagents, cost-aware model routing, and true wave parallelism. Ships as npx mindforge-cc.

Node.jsTypeScriptMCPsql.jsWebSocket

1,193 commitsDetails

Graph-Forge

AI-native distributed code-intelligence platform.

Models code as a Neo4j knowledge graph + Chroma embeddings over ~19 polyglot microservices, powering RAG over massive codebases with full observability.

GoPythongRPCNeo4jChroma

555 commitsDetails

Agent-Forge

Self-improving AI agent infrastructure.

A Karpathy-style propose/eval/score/commit-or-revert loop over a git-versioned agent spec, with an LLM-judge eval harness. Framework-agnostic.

PythonFastAPIpgvectorCeleryMCP

201 commitsDetails

Tombstone

Production intelligence layer for 5,000+ feature flags.

Self-hosted feature flag intelligence with blast-radius gating, circuit-breaker auto-rollback, causal incident correlation ("What Changed?"), and Knight Capital–style tombstoning. Go + Python + TypeScript polyglot.

GoPythonTypeScriptReactPostgreSQL

492 commitsDetails

Trelix

Offline code intelligence engine — Tree-sitter indexing, hybrid BM25+vector+graph search.

Zero-infra code intelligence — Tree-sitter AST indexing + 7-leg RRF retrieval (BM25 + vector + call-graph + SPLADE + multi-granularity) + knowledge graph (Louvain community detection, Pyvis). MCP server. 1,467 unit tests. Works offline with no API key.

Pythontree-sittersqlite-vecSQLiteBM25

501 commitsDetails

Inkforge

Notes, topics, or code → published technical articles.

STORM two-stage pipeline (outline → draft → polish) with BM25 in-memory RAG, streaming web UI, and one-command publishing to Dev.to, Hashnode, and a Playwright-rendered LinkedIn carousel.

TypeScriptNode.jsNext.jsTurborepoBedrock

89 commitsDetails

Order Processing System

Distributed, event-driven gRPC microservices.

A polyglot Go + Python/FastAPI order-processing backend — Saga orchestration, CDC outbox, etcd leader election, and metric-based replica read routing.

GoPythongRPCPostgreSQLKafka

Code Intelligence & EnginesDetails

View all 8 projects

// competitive programming & impact

Recognition

Google Code Jam 2023

AIR 420 · 3,687 / 85,000+

Meta Hacker Cup 2022

4,048 / 70,000+

Flipkart GRiD 2022

Top tier · 1,325 / 40,000+

Institute Rank 1

GeeksforGeeks & InterviewBit

Mentored 250–300 students

Data Structures & Algorithms

// writing

Latest articles

Dev.to

Jul 5, 2026

We Shipped Tombstone v1.0. Then We Found Nine Bugs That Would Have Paged Us at 2am.

What happened when we pressure-tested Tombstone v1.0 against production reality: a kill switch silently returning 400, approval endpoints wired to routes never registered, a scheduler that could double-fire. Nine bugs. Nine fixes. Here's the post-mortem.

devopsopen-sourcesystem-design

10 min

Substackalso on

Jul 5, 2026

I Built trelix Because I Was Tired of Grepping My Way Through Codebases

How I built trelix — a zero-infra code intelligence engine that indexes any repo with Tree-sitter, runs 7-leg RRF retrieval, and answers questions about your codebase offline. From four hours of archaeology to a query that returns in 200ms.

pythonopen-sourceai

12 min

All articles

// let's talk

Open to Backend, GenAI & Full-Stack roles

Building agent infrastructure or scaling an event-driven backend? I'd love to hear about it.

Schedule a call LinkedIn GitHub