In the community

Speaking at Conf42 Prompt Engineering 2026

Talk on keeping prompts stable as they change, using prompt contracts: structured outputs, explicit invariants, and automated tests.

Speaking at SF Tech Week 2026

Trust Issues: When Agents Act for You

Panelist at Trust Issues: When Agents Act for You, on identity, access, and authorization for AI agents.

HeapLens at SPLASH 2026

An IDE-integrated tool for heap dump analysis with in-editor source bridging and LLM-assisted diagnosis

Tool paper on HeapLens in the companion proceedings of SPLASH 2026, the ACM SIGPLAN International Conference on Systems, Programming, Languages, and Applications.

Evidence-Indexed Slide Narration

With heuristic checks and template fallback

Paper at the GroundLM 2026 workshop, co-located with EMNLP 2026, on keeping generated slide narration tied to what the slide actually contains.

• LLM Ops & Eval

Default Zero-Shot LLM APIs Over-Flag Indian Financial SMS

A policy-aligned benchmark and template-level pre-deployment audit

Paper at the GroundLM 2026 workshop, co-located with EMNLP 2026, on how often language-model APIs wrongly flag legitimate Indian financial text messages as fraud.

• LLM Ops & Eval

VERITYGATE

A four-gate schema-level faithfulness framework and paired benchmark for grounded LLM narrations over structured evidence

Paper at the GroundLM 2026 workshop, co-located with EMNLP 2026, on checking whether a model's written explanation actually follows the structured evidence it cites.

• LLM Ops & Eval

TRACE

Target-aware retrieval, attributed evidence, and contract-constrained extraction for LitTraceQA

Paper at the GroundLM 2026 workshop, co-located with EMNLP 2026, on the gap between finding the right paper and producing an answer a scorer can verify.

• RAG
• Search

EVICALC

Evidence first, arithmetic second: a system report and failure analysis for DocSem

Shared-task system paper at DocInsights 2026, co-located with EMNLP 2026, on reading a figure out of a PDF before computing with it.

• RAG

Program Committee member for IEEE BigData 2026 High School Symposium

Program Committee member for the IEEE BigData 2026 High School Symposium. Reviewed three student papers against a standard set for high school research.

Speaking at WeAreDevelopers

Test before release. Enforce at runtime.

Talk on governance for tool-using AI agents: why a correct answer can still be unsafe, the three checks that catch it in CI, and the runtime gate that stops it before the action runs.