Santa Clara, California • Agentic Systems & LLM Platforms (Adobe Brand Concierge)
Current Focus: Building LLM agent evaluation & optimization platforms, multi-tenant serving runtimes, grounded citation pipelines, and CPU ONNX retrieval services for Adobe Brand Concierge.
Biography & Background Santa Clara, CA
Rajasekhar Reddy Mekala
Santa Clara, California

I am a Software Engineer specializing in Agentic Systems & LLM Platforms at Adobe (Santa Clara, CA) on the core engineering team building Adobe Brand Concierge (after earlier backend, runtime-operations, and scaling work on Adobe Journey Optimizer). Joining roughly a month after initial product ideation, I have helped take Brand Concierge from early prototypes through enterprise production deployments across semiconductors & hardware, enterprise SaaS, retail & e-commerce, travel & hospitality, and sports & entertainment—serving as Primary Contributor of its LLM evaluation & optimization platform and building core subsystems across the conversation runtime, grounded citation pipeline, guardrails, and ONNX retrieval services.

Previously, I earned my M.S. in Computer Science at UC Irvine (UCI NLP Lab with Dr. Sameer Singh), researching in-context learning, prompt robustness, and chain-of-thought reasoning in LLMs (published at NAACL 2024 and EMNLP 2022), following a B.Tech at IIT Kharagpur. Earlier, I built distributed telemetry microservices at TruckX, search and analytics systems at Adobe (2018–2021), and cloud performance models at Adobe Research, co-authoring granted US Patent US11086646B2.

Adobe Brand Concierge — Agentic Platform, Evaluation & Retrieval Systems

Adobe Inc. • 2023–Present
PydanticAI • Model Context Protocol • GEPA Optimization • ONNX Neural Retrieval
Core platform engineering supporting global enterprise deployments across semiconductors, enterprise SaaS, retail, travel & hospitality, and sports/entertainment.
  • Agent Evaluation & Optimization Platform (Primary Contributor): Built the end-to-end offline and online evaluation platform (FastAPI, durable MSSQL staged store scaling to 50k-answer datasets, Okta SSO), integrating GEPA reflective prompt & judge optimization, multi-judge correlation diagnostics, guardrail analysis, offline source-attribution calibration, and replay evaluation runners.
  • Conversation & Agent Runtime Subsystems: Co-built the foundational conversation service from day one (FastAPI/MongoDB session & resource persistence, SSE streaming, A2A streaming client integration, tenant traffic/sandbox isolation, web-traffic analytics ingestion, and enterprise pilot integrations), and built declarative Pydantic/MCP tool schemas with pre/post hooks and configurable error-recovery policies, manifest-driven named LLM profile routing, document-first citation attribution, multi-layered preflight guardrails, and structured trace contracts in the successor PydanticAI runtime.
  • CPU ONNX Retrieval & Reranking Infrastructure: Extended the CPU Text Embeddings Inference (TEI) / ONNX Runtime serving stack with INT8 quantization pipelines, custom E5 cross-encoder and two-stage embedding build targets, and a FastAPI multi-query batch reranking gateway.
Selected Publications & Patents All Publications & Patents →
NAACL 2024 Selected EchoPrompt: Instructing the Model to Rephrase Queries for Improved In-Context Learning
Rajasekhar Reddy Mekala, Yasaman Razeghi, and Sameer Singh
Findings of the Association for Computational Linguistics: NAACL 2024
PDF arXiv:2309.10687
Abstract & BibTeX
We introduce EchoPrompt, a simple yet effective approach that prompts the language model to rephrase its queries before answering them. Just like humans, reminding the model of the question at hand by rephrasing it seems to help preserve context and reinforce connections across tokens. Through extensive experimentation with four causal model families across numerical reasoning (GSM8K, SVAMP), reading comprehension (DROP), and logical reasoning tasks, EchoPrompt yields substantial improvements (e.g., +5% on numerical reasoning and +13% on reading comprehension in Zero-Shot-CoT with code-davinci-002).
@inproceedings{mekala2024echoprompt,
  title={EchoPrompt: Instructing the Model to Rephrase Queries for Improved In-Context Learning},
  author={Mekala, Rajasekhar Reddy and Razeghi, Yasaman and Singh, Sameer},
  booktitle={Findings of the Association for Computational Linguistics: NAACL 2024},
  year={2024},
  url={https://arxiv.org/abs/2309.10687}
}
EMNLP 2022 Selected Snoopy: An Online Interface for Exploring the Effect of Pretraining Term Frequencies on Few-Shot LM Performance
Yasaman Razeghi*, Raja Sekhar Reddy Mekala*, Robert L. Logan IV, Matt Gardner, and Sameer Singh
Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing: System Demonstrations (EMNLP 2022)
PDF ACL Anthology
Abstract
SNOOPY is an online interface for exploring Pile term frequencies alongside precomputed model performance across NLP benchmarks. It lets users inspect how term frequency relates to few-shot accuracy across models, example counts, and datasets.
US Patent Selected US11086646B2: Tenant-side detection, classification, and mitigation of noisy-neighbor-induced performance degradation
Subrata Mitra, Sopan Khosla, Sanket Vaibhav Mehta, Rajasekhar Reddy Mekala, and Aashaka Dhaval Shah
Granted US Patent 11,086,646 (Adobe Inc., Aug 2021; published as US20190354388A1)
JOSS 2020 SunPy: A Python package for Solar Physics
Stuart J. Mumford et al., including Rajasekhar Reddy Mekala
Journal of Open Source Software, 5(46), 1832 (2020)
Core Engineering Pillars
Career & Education Snapshot
Ecosystem & Profiles