Vetted Latency Optimization Professionals

“Built and shipped production RAG/agentic systems in high-stakes domains (biomedical and legal), including an enterprise biomedical document retrieval platform over ~10k scientific docs and a multilingual African-law assistant at the World Bank. Deep hands-on experience with LangChain/LangGraph/LlamaIndex and evaluation tooling (LLM-as-a-judge, safety/hallucination detection), with measurable gains in retrieval quality and hallucination reduction.”

“University hackathon finalist (2nd place) who built CareerSpark, a production-style multi-agent career guidance app in 24 hours using a hierarchical debate architecture with a moderator/judge agent. Has startup internship experience at LiveSpheres AI using LangChain for multi-LLM orchestration, and demonstrates a structured approach to testing/evaluation (golden sets, integration sims, latency/accuracy KPIs) plus strong non-technical stakeholder communication.”

“Recent J&J intern who built a conversational RAG agent and led a shift from a monolithic model to a modular RAG workflow, cutting response time from several days to under a second by tackling data fragmentation, context retention, and embedding/latency optimization. Also worked on a large (7B-parameter) multimodal VQA pipeline for healthcare research and stays current via NeurIPS/ICLR and open-source contributions.”

“Backend engineer with American Express experience maintaining an internal Python/Flask rewards simulation microservice used by product analysts and QA. Demonstrated strong performance and scalability work: moved batch simulations to Celery, added Redis caching to cut DynamoDB latency, and tuned Postgres/SQLAlchemy queries with EXPLAIN ANALYZE and composite indexes (bringing API responses under ~200ms by queueing jobs). Also has experience integrating ML via Flask-based model-serving APIs (scikit-learn/LightGBM packaged with joblib) and designing multi-tenant data isolation and tenant-specific configuration systems.”

“Built and shipped a production LLM-powered RAG system at Upstart enabling natural-language search across 50k+ scattered internal technical docs. Delivered sub-300ms p95 latency for ~50 active users with strong hallucination safeguards (retrieval-first, thresholds, citations) plus robust testing/monitoring and cost controls (prompt caching cutting API spend ~20%).”

“Built and shipped a production RAG-based assistant that lets parents ask natural-language questions about their child’s learning progress, using pgvector retrieval (child-id filtered) and Redis caching to hit ~180ms latency. Implemented real-world guardrails and compliance (Llama Guard, COPPA, retrieval thresholds, fallbacks) with 99.5% uptime, and ran human-in-the-loop eval loops that improved satisfaction from 3.8 to 4.2 while serving 60k+ monthly users and reducing costs significantly.”

“Built and productionized an enterprise retrieval-augmented generation platform for internal knowledge over large unstructured corpora, emphasizing trust via strict citation/grounding and hybrid retrieval (BM25 + FAISS + cross-encoder re-ranking). Demonstrates strong scaling and cost/latency optimization through incremental indexing/embedding and index partitioning, plus disciplined evaluation/observability practices. Has experience operationalizing pipelines with Airflow/Databricks/GitHub Actions and partnering closely with risk & compliance stakeholders on auditability requirements.”

“Built and deployed a production SecureAIChatBot (RAG-based) for secure internal information retrieval, using embeddings/vector search, GPT models, monitoring, and safety filters. Focused on real-world production challenges like latency and output consistency, applying caching, retrieval scoping, smaller models, and controlled prompting, and used LangChain to orchestrate the end-to-end workflow.”

“Built a production, real-time clinical documentation system at HCA that converts doctor–patient conversations into structured clinical summaries using speech-to-text, LLM summarization, and RAG. Demonstrated measurable gains from medical-domain fine-tuning (clinical concept recall +18%, ROUGE-L 0.62 to 0.74) while meeting HIPAA constraints via PHI anonymization and encryption, and deployed via Docker/FastAPI with CI/CD and monitoring.”

“Robotics software engineer focused on system-level C++/Linux stacks for multi-robot platforms, owning the communication layer and validation/testing infrastructure. Built Python simulation/replay and fault-injection tooling integrated with Docker + GitLab CI/CD, and debugged real-time localization issues by instrumenting IPC timing and refactoring multi-threaded pipelines for deterministic performance.”