We don't have the transmit portion yet done, will Lux be putting their transmit element in the initial samples?  

My take on this Baba. Short answer is yes. But let me explain.

If we go back in time. Some of you may recall that Suresh was of the opinion that the module makers were not interested in just half a solution. They did not want to change half of the module but would prefer to wait for the Transmit and Receive optical engines. Then as the world turns the 800G pluggable becomes available using existing technologies but at a cost. Suresh recently talked about how that market is being served today. He called the 800G pluggables available today a work of art. Every bit of space is used to fit all the components required applying existing technologies commercially available today. 

So as it turns out there is a demand for POETs highly integrated and efficient 400G/800G RX solution as a matter of necessity. Providing lower cost, increased power and thermal efficiency. and at the same time freeing up needed space within the QSFPDD footprint to improve the transmit side layout. It's transitional as they wait for the POET TXRX engines for 800G.

The Poet 400G FR4 RXOE, produced by Super Photonics Xiamen, the company's joint venture, integrates a quad transimpedance amplifier (TIA), four high-speed photodiodes and an optical demultiplexer integrated in a single chiplet measuring 12 millimetres by five mm.

As a reminder Suresh has advised us that he intends to demonstrate a game changing 800G optical engine at OFC. I think we are in for some fireworks.

The blog does a really good job at explaining this : POET Technologies (poet-technologies.com)