Ananda Poudel

Platform engineer building enterprise agentic AI on open-source infra.

Eight years building backend and platform systems in production. Most recently I built an enterprise agentic-AI platform from scratch — model routing, sandboxed execution, agent memory, and the observability and guardrails that keep autonomous workflows safe at 500+ requests per second. Most of it runs on open source; I treat agent infrastructure as a distributed-systems problem, not an API wrapper.

▸ Open to relocation — preferring Seattle or the Bay Area

Email me What I bring ▸

500+req/s agentic platform
8 yrsshipping production systems
−80%unverified tool calls

What I bring

Senior platform engineer since 2018, remote. Three areas where I do my strongest work — with the numbers behind them.

Agentic AI Platform

500+ req/s · model routing · sandboxed execution · memory · tracing

Built the platform end to end for 500+ concurrent requests/second — LiteLLM as the orchestration gateway, routing each task to the model that fits on latency and cost and standardizing how agents call external tools.
Designed the execution layer on Kubernetes with E2B — ephemeral sandboxes for autonomous workflows, with agent code isolated from the host network by default.
Built two-tier agent memory: Redis for short-term session state, Milvus for long-term vector storage — long-running workflows keep context and resume across sessions.
Instrumented the reasoning loop with OpenTelemetry into Langfuse, making every decision step inspectable (both open source).
Layered Guardrails AI over tool execution to enforce programmatic policy — unverified tool calls dropped over 80%, which is what made autonomous workflows safe to run.

LiteLLM
E2B
Kubernetes
Milvus
Redis
Langfuse
OpenTelemetry
Guardrails AI

LLM Tooling & Developer Experience

Definitions → production pipeline code, with evals gating every change

Own the LLM-based compilation tooling that turns internal modeling definitions into executable pipeline code — prompt construction, context assembly, structured-output parsing, and a validation layer that gates anything reaching production.
Built its evaluation harness — rubric scoring plus human review against a held-out regression set on every prompt or model change.
Introduced LLM-assisted code review and ran the prompt-engineering training — review turnaround down ~25% with no drop in defect catch rate.

Structured Outputs
Evaluation Harnesses
Prompt Engineering
Python
Developer Tooling

Backend & Infrastructure

Six services · ~800k req/day · 99% availability · −300ms P99

Build and operate the REST & GraphQL services other teams build against — across Node/Express, Go, and C#/.NET Core with async event processing; six core services carrying ~800k requests/day for four product teams, with contract versioning.
Built the hybrid on-prem→cloud data layer — ETL moving multi-TB datasets in near real time at 99.9% accuracy, the foundation for analytics and ML.
Cut P99 API latency 300ms and raised peak throughput 50% (SQL tuning + Redis caching); SLO monitoring and distributed tracing hold 99% availability.
Own CI/CD and Bicep IaC — zero-downtime deployments, ~40% less manual infra work, and managed identities + Key Vault that removed hardcoded secrets entirely.
Run the RFC and design-review process that keeps architecture aligned across teams, and mentor junior engineers.

Go
Node/Express
C# / .NET
GraphQL
Kafka
Azure
Bicep
CI/CD

Inventory

Agentic AI & LLM Infra

LiteLLM
E2B
Milvus
Langfuse
Guardrails AI
OpenTelemetry
Agentic Orchestration
Model Routing & Fallback
Agent Memory
RAG
Embeddings & Semantic Search
Prompt Engineering
Structured Output Validation
Evaluation Harnesses
Token & Cost Monitoring
Anthropic & OpenAI APIs

Languages & Backend

Go
Python
TypeScript
Node.js / Express
C#
.NET Core
ASP.NET Web API
SQL / T-SQL
GraphQL
gRPC
REST
Microservices
Event-driven

Infrastructure & Cloud

Kubernetes
Docker
Redis
Kafka
PostgreSQL
Linux
Git
GitHub Actions
Azure (AKS, App Insights, Key Vault)
Managed Identities
Bicep
Azure DevOps
CI/CD
Zero-downtime Deploys
Observability & Tracing

Data & Frontend

ETL Pipelines
Data Lake Architecture
Stream Processing
Real-time Analytics
React
Next.js
Redux
React Native

Records

Education

B.S. Software Engineering

Saint Cloud State University · GPA 3.9 · 2016–2019

Publication

“Interior Design with AR”

Published in MICS proceedings.

Contact

Open to Senior / Staff platform & software-engineering roles — open to relocation, preferring Seattle or the Bay Area.

email@anandapoudel.com

☎ PHONE — LOCKED

Catch 8 coins to unlock — move with mouse, finger, or ← →.

COINS 0/8READY

Open to relocation · Seattle / Bay Area preferred