Sandeep Malik

About

I'm an AI Engineer and Data Architect with 13+ years across 11 companies and 2 continents. Today I lead a central team at Autodesk building an agentic-AI platform — a BYO-agent / BYO-MCP model with its own discovery, orchestration, and harness/standards layers — and I've shipped 3 production agents, 5 MCPs, and 2 IDE plugins used across the company. One flagship: a proactive agent that modernized 1,500+ internal apps, migrating legacy, vulnerability-prone APIs to GraphQL across five languages at >85% success — with strong ROI.

I'm a hands-on builder who takes ideas from MVP and POC to production-grade systems — agents, MCP tools, and RAG — on top of a decade of data-platform and big-data architecture. Along the way I've handled ~2 PB of data and ~200K events/sec, built Databook's first semantic layer from scratch, and architected platforms at Cloudera, Sitecore, Intact, Ericsson, and CGI.

I care about systems that are reliable as well as smart — data integrity, observability, and evals matter as much as embeddings and prompts. I've led teams of engineers, because great platforms are built by great teams.

Projects & Highlights

Featured · Open Source

Agentic Expense Manager

Local-first income & expense tracker driven by a Claude tool-calling agent — chat it, WhatsApp it, or snap a receipt. SQLite is the single source of truth; OCR and Google sync are opt-in.

One agent, two channels — web chat and WhatsApp share the same service layer.
Server-side tax back-calculation reused by REST, the agent, and live previews.
200+ backend tests, including an adversarial suite built to break reconciliation.

PythonReactClaude AgentSQLiteDocker

View on GitHub →

Agentic Platform · Autodesk

Leading a BYO-agent/MCP platform — agent & MCP catalog, discovery, orchestration, and worker standards. Shipped 3 agents, 5 MCPs & 2 IDE plugins used across Autodesk.

AgentsMCPHarness

API Modernization Agent · Autodesk

A proactive agent that migrates legacy, vulnerability-prone APIs to GraphQL across 1,500+ internal apps — direct & dependency-based usage in Python, Go, Java, Rails & Perl — at >85% success by code complexity, with a UI dashboard. High ROI.

Proactive AgentGraphQL1,500+ apps

Databook Semantic Layer

Built Databook's data platform from scratch and its first semantic layer (early 2023) powering Databook AI — with agentic workflows in LangGraph & Google ADK.

LangGraphSemantic Layer

GPT Sales Product · Advanced RAG

A GPT product for strategic sales with multi-query & step-back retrieval, reranking, and LLM-as-judge to cut hallucinations and lift engagement.

RAGPinecone

Search at Scale · Sitecore

Led the core Search team handling ~200K events/sec; architected a Data Lake/Warehouse for a global retailer, cutting storage costs.

ElasticsearchKafkaSnowflake

EN-RuleEngine

A Scala rule engine that validates streaming and batch data against rules defined in YAML — data-quality enforcement for pipelines.

ScalaStreaming

View on GitHub →

Experience

Jul 2025 — Present · Montreal

Principal Software Engineer / Tech Lead

Autodesk · Trust AI Engineering

Lead a central team building an agentic platform (BYO-agent/MCP) — agent & MCP catalog, discovery and orchestration layers, and the harness/standards for building workers.
Shipped 3 production agents (1 proactive, 2 reactive), 5 MCPs, and 2 IDE plugins used across Autodesk.
Built a proactive agent that migrates legacy, vulnerability-prone APIs to GraphQL across 1,500+ internal apps — direct & dependency-based usage in Python, Go, Java, Rails & Perl — at a >85% success rate (by code complexity), surfaced via a UI dashboard; strong ROI.
Shift security left: MCP tools surface secure-coding standards inside AI IDEs; remediation agents auto-detect repo vulnerabilities and raise fix PRs.

Dec 2022 — Jul 2025 · Montreal

Data Architect / Staff Engineer

Databook

Built the data platform from scratch and its first semantic layer (early 2023) powering Databook AI.
Spearheaded a GPT sales product with advanced RAG (multi-query, step-back, reranking, LLM-as-judge) and agentic workflows in LangGraph & Google ADK.
Semantic index over Pinecone & Elasticsearch on Snowflake/Databricks; led a team of data engineers.

Mar 2022 — Dec 2022 · Montreal

Principal Data Engineer

Sitecore

Led the core Search team handling ~200K events/sec — AI-backed search and personalization for e-commerce.
Architected a Data Lake/Warehouse for a global retailer, reducing storage costs; built Snowflake marts and automated monitoring.

May 2021 — Mar 2022 · Montreal

Solutions Architect

Cloudera

Professional Services expert onboarding customers to Cloudera Data Platform (public/private/hybrid); designed on-prem → AWS migrations and trained engineering teams.

Apr 2020 — May 2021 · Montreal

Senior Data Engineer

Intact Insurance

Led a team of data engineers; built on-prem → AWS historical-data migration and refactored pipelines onto Databricks.

2012 — 2020 · Canada, Dubai & India

Data / Software Engineer

Ericsson · CGI · Synechron · Interglobe (Dubai) · MothersonSumi · NIIT

A decade of data & software engineering — 5G optimization, real-time trade-settlement pipelines, an in-memory data fabric, batch + real-time big-data over ~2 PB (Dubai), and enterprise platforms.

Education & Certifications

degree

M.Sc. Computer Science

Master of Science in Computer Science — Manav Bharti University (2013).

certifications

Cloud & Data

Snowflake SnowPro Core · AWS Solutions Architect – Associate · AWS Cloud Practitioner · Databricks Spark 3.0 · CCA-175 · Microsoft AZ-900.

focus

Harness & Agentic Engineering

Building the harness around LLMs — agent runtimes, MCP, orchestration, evals, and guardrails for production-grade agentic systems.