1. AI Platform Engineer | LLM Systems Architect
2. OVERVIEW
3. AI Platform Engineer building production-grade LLM systems with strong
4. reliability guarantees.
5. 10+ years in distributed systems and high-availability infrastructure, now
6. focused on AI control planes, RAG pipelines, routing, evaluation frameworks,
7. and observability.
8. Designed LLM orchestration systems with explicit validation, retry, rollback,
9. and commit semantics to reduce non-determinism and enforce correctness. Brings
10. deep experience in scaling, failure modeling, and SLO-driven production system
11. design.
12. SKILLS
13. AI Platform & LLM Systems: RAG architectures, LLM orchestration, control
14. planes, routing & fallback logic, evaluation frameworks (Langfuse, deepeval),
15. trace-level observability, prompt/version management, context engineering,
16. agentic workflows
17. Distributed Systems & Infrastructure: High-availability design, autoscaling
18. concepts, failure modeling, SLO/SLA enforcement, load balancing
19. (HAProxy/Nginx), Kubernetes, Docker, cloud-native architectures
20. (AWS/GCP/Azure), Rackspace, Terraform, Ansible, Chef, Linux, Solaris, ZFS
21. Systems Design & Leadership: Architecture Reviews, Design RFCs, Failure
22. Modeling, Incident Postmortems, Cross-team platform enablement
23. Languages: Python, SQL, Bash
24. Databases: PostgreSQL (expert: performance tuning, query optimization,
25. replication, partitioning, migrations, security), MySQL (strong: performance
26. tuning, optimization), CockroachDB, Cassandra, MS SQL Server, Oracle
27. EXPERIENCE
28. NetApp (Acquired), Remote — Senior Systems Architect
29. May 2022 - PRESENT
30. ●
31. Designed and deployed PostgreSQL-backed LLM platform architectures using
32. RAG and Model Context Protocol (MCP), treating the database as an AI
33. control plane rather than a passive datastore. Improved internal
34. operational workflows and reduced failure rates.
35. ●
36. Built and productionized multi-agent LLM pipelines using LangGraph,
37. modeling workflows as explicit state machines with validation, retry,
38. correction, and commit phases — increased reliability and contributed to
39. ~2x YoY revenue growth.
40. ●
41. Integrated end-to-end LLM observability and evaluation pipelines
42. (Langfuse, deepeval), enabling trace-level introspection, automated
43. scoring, regression detection, and feedback-driven iteration.
44. ●
45. Defined system invariants, routing gates, and validation probes to
46. enforce correctness, reproducibility, and safe deployment of AI-assisted
47. database tooling in enterprise SaaS environments.
48. ●
49. Applied distributed systems principles (idempotency, transactional
50. boundaries, rollback semantics, auditability) to AI workflows, reducing
51. silent failures and improving deployment safety.
52. ●
53. Designed API-layer abstractions and LLM gateway components to support
54. routing, validation, and model versioning strategies.
55. ●
56. Deployed & iterated LLM systems in cloud-native environments (AWS/Azure),
57. integrating storage, compute, and monitoring layers to support production
58. workloads.
59. Instaclustr (Acquired), Remote — Senior Database Reliability
60. Engineer
61. March 2021 - May 2022
62. ●
63. ●
64. ●
65. PostgreSQL scalability expert for DoorDash, helping them scale their
66. Postgres RDS & Aurora systems to support rapid traffic growth during the
67. COVID-19 pandemic surge.
68. Led the team of database experts working on all aspects of DoorDash
69. database scalability, performance, and infrastructure overhaul
70. Provided ongoing schema design, query optimization, live production
71. schema change processes with minimal disruption, and disaster recovery
72. consulting playing a pivotal role in the company’s growth & operations
73. leading to a successful IPO
74. Credativ (Acquired), Remote — Senior Database Reliability
75. Engineer
76. January 2019 - March 2021
77. ●
78. PostgreSQL & MySQL specialist for enterprise clients including DoorDash,
79. Etsy, National Geographic, Follow Up Boss, Gilt, Leafly, and Paperless
80. Post among others, maintaining up to 99.999% uptime for mission-critical
81. operations
82. ●
83. Led large-scale PostgreSQL schema changes, migrations, major upgrades,
84. security patching, and performance optimization in high-transaction,
85. high-availability environments
86. ●
87. ●
88. 24/7 oncall - Prevented critical outages during peak traffic through
89. proactive tuning, capacity planning, and failure modeling, contributing
90. to long-term contract renewals
91. Conducted comprehensive database performance & security audits for
92. numerous clients providing surgical, concrete solutions to complex issues
93. & pain points
94. OmniTI Computer Consulting, Remote — Data Engineer
95. June 2013 - January 2019
96. ●
97. Re-designed OLAP ETL pipeline and data sanitization project for Gilt
98. Japan saving them $2+ million per year on infrastructure and resources
99. costs
100. ●
101. ●
102. ●
103. ●
104. Designed and implemented HA/DR strategies that reduced processing time by
105. 65% and recovery time objectives (RTO) by 70%
106. 24/7 PagerDuty oncall for multiple clients - debugging and solving
107. production critical performance and availability issues at all times of
108. the night
109. Built and contributed to several open source projects for database
110. monitoring, security, replication, observability, reporting,
111. partitioning, sharding, and migration and employed these tools to better
112. serve clients
113. Performed in-depth & thorough configuration tuning for Postgres and MySQL
114. as well as Linux & Solaris OS to improve database performance and
115. efficiency for multiple clients. Clients like Locally and Click2Ship
116. experienced drastic improvements (4x traffic with no issues) in scaling
117. for holidays & random spikes as a direct result of my endeavors
118. SELECTED PROJECTS
119. ●
120. ●
121. Enterprise AI platform for automated DBMS migration and code conversion
122. using multi-agent orchestration with full auditability, evaluation
123. pipelines, and feedback loops. Designed for safe migration of
124. heterogeneous workloads under strict correctness and traceability
125. guarantees.
126. Designed and implemented RAG + MCP financial analytics platform with
127. routing logic, persistence layer, and evaluation-driven pipelines for
128. reproducible AI-assisted analysis.
129. ●
130. ●
131. Built real-time streaming RAG + MCP pipeline as proof-of-concept for
132. reliable LLM deployment under latency constraints, integrating
133. observability and controlled rollout mechanisms.
134. In-memory enterprise data sanitization pipeline for PCI-compliant
135. PostgreSQL environments
136. CONFERENCE TALKS
137. Spoke at SCaLE 16x, 18x, 23x (most recent), pgCon.dev (Ottawa), Percona Live,
138. NYCPUG, CPOSC on various topics:
139. ●
140. ●
141. ●
142. ●
143. ●
144. Postgres as an AI Control Plane: Building RAG + MCP Workflows Inside the
145. Database (2026)
146. Provisioning & Automating High Availability Postgres on AWS
147. Securing Your Data on Postgres
148. Data Mining for Beginners with Pandas
149. Think Your Postgres Backups & Disaster Recovery Are Safe? Let’s Talk.
150. EDUCATION
151. University of Maryland Baltimore County, Baltimore County MD —
152. Masters in Computer Science
153. September 2011 - December 2014
154. ●
155. Research Assistant for NSF funded massive data processing & visualization
156. project for geologists
157. Panjab University, Chandigarh (India) — Bachelor of Engineering
158. in Computer Science
159. July 2007 - June 2011
160. CERTIFICATIONS
161. AWS Certified Solutions Architect
162. AWS Certified Cloud Practitioner
163. TECHNICAL BLOGS & LINKS
164. https://medium.com/@reliable-by-design
165. https://penningpence.blogspot.com
166. https://github.com/payals