.png){kind=logo}
Harsha Sikha
Mid-level AI/ML Engineer specializing in Generative AI and data engineering
Connect with Harsha
Harsha already has a relationship with Reval, so a warm intro from us gets a much better response than cold outreach.
Recommended
Already have an account?
About
IBM engineer who built and deployed a production RAG-based LLM assistant using LangChain/FAISS with a fine-tuned LLaMA model, served via FastAPI microservices on Kubernetes, achieving 99%+ uptime. Demonstrates strong practical expertise in reducing hallucinations (semantic chunking + metadata-driven retrieval) and managing latency, plus mature MLOps practices (Airflow/dbt pipelines, MLflow tracking, monitoring, A/B and shadow deployments) and effective collaboration with non-technical stakeholders.
Experience
Education
Key Strengths
- Built and deployed production RAG LLM assistant at IBM
- Reduced hallucinations via semantic chunking and metadata-filtered/boosted retrieval
- Latency optimization through vector index tuning
- Production reliability focus (monitoring, drift/confidence tracking, health checks)
- Experimentation rigor (offline benchmarks, A/B tests, shadow deployments)
- End-to-end data pipeline orchestration for RAG using Airflow + dbt
- Strong cross-functional collaboration translating business goals into AI requirements
- Communicates AI progress in business outcomes (fewer escalations, faster resolution)
- Achieved 99%+ uptime in production
Discover more candidates like Harsha
Search across thousands of pre-screened, high-quality, high-intent candidates on Reval.
Search Talent