Vetted PySpark Professionals

“Developed and productionized VL-Mate, a vision-language, LLM-powered assistant aimed at helping visually impaired users understand their surroundings and query internal knowledge. Emphasizes reliability and safety via confidence thresholds, uncertainty-aware fallbacks, hallucination grounding checks, and rigorous offline + user-in-the-loop evaluation, with experience orchestrating multi-step LLM pipelines (LangChain-style and custom Python async) and deploying on containerized infrastructure.”

“Built a production lead-generation system using AI agents that researches the internet for relevant leads and integrates RAG-based contact enrichment/shortlisting aligned to existing CRM data, enabling sales reps to focus more on selling. Also has hands-on AWS data orchestration experience (Glue, Step Functions) moving raw data into Redshift and evaluates agent performance with human-in-the-loop plus BLEU/perplexity metrics.”

“Full-stack/ML-leaning engineer who has shipped production-grade real-time analytics and an internal AI support assistant using RAG over enterprise documentation. Demonstrates strong systems thinking across scalability, reliability, observability, and LLM safety/evaluation (thresholded retrieval, RBAC, response validation, regression-gated evals), with concrete iteration based on performance metrics and user feedback.”

“Built multiple systems from scratch at DSSD and Aglint, including an NGO sustainability reporting dashboard and a production LLM-powered phone screening agent using Twilio/Retell AI with RAG grounded in PostgreSQL candidate/job data. Strong focus on real-world reliability: guardrails, monitoring, and lightweight eval/regression loops that reduced recruiter score overrides by ~30%. Currently on OPT through May 2026 (plans STEM OPT extension) and committed to relocating to NYC for in-person work; seeking $90k–$120k base with meaningful equity for founding engineer roles.”

“Built a production Apple-focused LLM Q&A bot that answers user issues using similar past discussion records, including large-scale scraping and cleaning of thousands of forum threads. Used BeautifulSoup + Playwright for static/dynamic extraction, PySpark + NLP for preprocessing, and LangChain RAG with a custom response-likeliness metric to evaluate performance.”

“Founding engineer at a startup building LLM/agentic workflows for public-safety customers, with hands-on experience delivering a hybrid on-prem + secure cloud solution to meet strict compliance needs. Implemented OpenTelemetry observability for multimodal agentic systems behind closed networks and used the resulting traces to optimize prompting/token usage for customer-specific security integrations. Regularly runs technical workshops and supports pre/post-sales by translating integration feedback into product roadmap decisions.”

“At Provenbase, built and shipped a production LLM-powered semantic search and candidate matching platform (RAG with GPT-4/Gemini, multi-agent orchestration, Elasticsearch vector search) to scale sourcing across 10M+ candidate records and 1000+ data sources. Drove sub-second performance, cut LLM spend 30% with routing/caching, and improved recruiting outcomes (+45% sourcing accuracy; +38% visibility of underrepresented talent) through bias-aware ranking and tight collaboration with recruiting stakeholders.”

“Backend/data engineer who built a production analytics data service (Python/FastAPI on AWS/Postgres with PySpark ETL) handling millions of records per day and drove major latency improvements (10–15s to <2s) via indexing, Redis caching, and shifting aggregations into ETL. Also shipped an LLM-based natural-language-to-SQL assistant end-to-end with strong guardrails (schema restrictions, read-only validation, RBAC, masking) and designed a multi-step agent workflow with verification and fallback logic.”

“Backend engineer with fintech/transaction-processing experience who built and optimized a Spring Boot + PostgreSQL + AWS service handling money transactions, resolving peak-traffic latency via query/index and connection pool tuning. Shipped an LLM-driven risk-flagging workflow integrated via a FastAPI Python service, owning prompt design, validation guardrails, monitoring, and human-in-the-loop escalation to reduce false positives and improve precision over time.”

“Built and productionized a domain-specific LLM-powered RAG knowledge assistant at JerseyStem for answering questions over large internal document corpora, owning the full stack from FAISS retrieval and LoRA/QLoRA fine-tuning to AWS autoscaling GPU deployment. Drove measurable gains (28% accuracy lift, 25% latency reduction) and improved reliability through hybrid retrieval, grounded decoding, preference-model reranking, and Airflow-orchestrated pipelines (35% faster runtime), while partnering closely with non-technical stakeholders to define success metrics and ensure adoption.”